Our research proposes two different training and modelling approaches of Generative Pre-trained Transformers (GPT) with specialized news feeds specific to the Spanish market: in-context example prompts and fine-tuned GPT models. Our findings indicate that integrating GPT insights into electricity price trend forecasting can result in more precise predictions and a deeper understanding of market dynamics. Through our research, we aim to provide insights to understand the capabilities of GPT solutions, and how those can be used to enhance prediction accuracy, ultimately supporting informed decision-making for stakeholders across the Spanish electricity market and companies whose margins heavily depend on electricity costs and price volatility.

In a densely populated country like Bangladesh, universities grapple with the challenge of efficiently addressing myriad queries from a large student body, leading to a heightened workload for university stakeholders. To tackle these challenges, we introduce PixieGPT, a tailor-made Generative Pre-Trained Transformer for Bangladeshi universities. PixieGPT significantly mitigates workload by adeptly handling common university-related queries, thereby enhancing user experience. The hierarchical structure plays a crucial role in managing diverse queries from thousands of students about the university system. The solution introduces a modular hierarchical knowledge base (KB) with simpler complexities, addressing the intricacies of efficiently managing large volumes of queries. PixieGPT is designed in a modular way so that the solution is also adaptable for the implementation of other universities worldwide based on the requirements of a particular administrative system. The modular nature facilitates easy adaptation with minor changes based on specific university requirements, ensuring a seamless integration process. This paper delves into the intricacies of PixieGPT's design, emphasizing its pivotal role in mitigating workload challenges for university stakeholders in Bangladesh. The incorporation of BERT for Natural Language Understanding(NLU) and GPT models for Natural Language Generation(NLG) enhances PixieGPT's capabilities, contributing to the scalability and efficiency of the system. The presented use case underscores the practical benefits of PixieGPT, positioning it as a promising solution for universities globally with similar operational frameworks.

One of the most burning topics in cybersecurity in 2023 will undoubtedly be the compliance with the Software Bill of Materials. Since the US president issued the Executive Order 14028 on Improving the Nation’s Cybersecurity, software developers have prepared and bills are transmitted to vendors, customers, and users, but they don’t know what to do with the reports they are getting. In addition, since software developers have identified the values of the Software Bill of Materials, they have been using the reports extensively. This article presents an estimate of 270 million requests per month, just from form one popular tool to one vulnerability index. This number is expected to double every year and a half. This simple estimate explains the urgency for automating the process. We propose solutions based on artificial intelligence and machine learning, and we base our tools on the existing FAIR principles (Findable, Accessible, Interoperable, and Reusable). This methodology is supported with a case study research and Grounded theory, for categorising data into axis, and for verifying the values of the tools with experts in the field. We showcase how to create, and share Vulnerability Exploitability eXchange data, and automate the Software Bill of Materials compliance process with AI models and a unified computational framework combining solutions for the following problems: (1) the data utilisation problem, (2) the automation and scaling problem, (3) the naming problem, (4) the alignment problem, (5) the pedigree, and provenance problem, and many other problems that are on the top of mind for many security engineers at present. The uptake of these findings will depend on collaborations with government and industry, and on the availability and the ease of use of automated tools.

Background: Diabetes mellitus (DM) and osteoporosis are two of the most widespread metabolic disease in the world. The aim of this study is to investigate the prevalence of DM among patients affected by osteoporosis and fragility fractures, and to search differences in clinical features. Methods: 713 patients belonging to CTO Bone Unit because of osteoporosis and fragility fractures were divided into two groups, according to the diagnosis of DM. The clinical and bone characteristics of patients were compared. Results: Prevalence of DM was 10.8%. When compared to patients without DM, the median age at the time of first fracture was similar (72 years ± 14 IQR vs 71 years ± 12 IQR); prevalence of combination of vertebral and hip fractures was higher (p=0.039), as well as prevalence of males, although not significantly (p=0.051). Bone mineral density (BMD) at all sites was higher in DM group; Trabecular Bone Score (TBS), instead, was significantly lower (p<0.001). Conclusions: Patients with fragility fractures and DM more frequently show combination of major fractures with higher BMD levels. In these patients, TBS could be a better indicator of bone health than BMD and therefore should be used as a diagnostic tool in clinical practice.

Pre-trained Language Models (PTLM) have remarkable and successful performance in solving lots of NLP tasks nowadays. And previous researchers have created many SOTA models and these models are included in many long surveys(Qiu et al., 2020). So, we would like to conduct a simple and short survey on this topic to help researchers understand the sketch of PTLM more quickly and comprehensively. In this short survey, we would provide a simple but comprehensive review of techniques, benchmarks, and methodologies in PTLM. And we would also introduce the applications evaluation of PTLM in this simple survey.

China's economic development has shifted from high speed to high quality. This study constructs an evaluation index for China’s high-quality development utilizing a new development concept. The entropy method and cluster analysis were used to assess the level of high-quality development in China between 2015 and 2021. Kernel density estimation method was employed to explore spatiotemporal dynamics. The results reveal significant differences in the level of economic development among provinces in China; the eastern region is significantly better than the central, northeastern, and western regions. From the perspective of the sub-item evaluation, challenges with uneven, uncoordinated, and insufficient economic development exist in various Chinese provinces.

Several researchers have contemplated deep learning-based post-filters to increase the quality of statistical parametric speech synthesis, which perform a mapping of the synthetic speech to the natural speech, considering the different parameters separately and trying to reduce the gap between them. The Long Short-term Memory (LSTM) Neural Networks have been applied successfully in this purpose, but there are still many aspects to improve in the results and in the process itself. In this paper, we introduce a new pre-training approach for the LSTM, with the objective of enhancing the quality of the synthesized speech, particularly in the spectrum, in a more efficient manner. Our approach begins with an auto-associative training of one LSTM network, which is used as an initialization for the post-filters. We show the advantages of this initialization for the enhancing of the Mel-Frequency Cepstral parameters of synthetic speech. Results show that the initialization succeeds in achieving better results in enhancing the statistical parametric speech spectrum in most cases when compared to the common random initialization approach of the networks.

The log data generated during the operation of a software system contains information about the system, and using logs for anomaly detection can detect system failures in a timely manner. Most of the existing log anomaly detection methods are specific to a particular system, have cold start problems, and are sensitive to updates in log format. In this paper, we propose a log anomaly detection method LogBD based on pre-training model and domain adaptation, which uses the pre-training model BERT to learn the semantic information of logs, and can solve the problems caused by multiple meaning words and log statement updates. An anomaly detection method LogBD based on distance to determine anomalies is constructed based on domain adaptation, using TCN networks to extract common features of different system logs and map them to the same hypersphere space. Finally, experiments are conducted on two publicly available datasets to evaluate the method, and the experimental results show that the method can better solve the log instability problem and has some improvement in the cross-system log anomaly detection effect.

This paper presents our approaches for the SMM4H’24 Shared Task 5 on the binary classification of English tweets reporting children’s medical disorders. Our first approach involves fine-tuning a single RoBERTa-large model, while the second approach entails ensembling the results of three fine-tuned BERTweet-large models. We demonstrate that although both approaches exhibit identical performance on validation data, the BERTweet-large ensemble excels on test data. Our best-performing system achieves an F1-score of 0.938 on test data, outperforming the benchmark classifier by 1.18%.

Smartphones are prone to SMS phishing due to the rapid growth in the availability of smart mobile technologies driven by Internet connections. Also, detecting phishing SMS is a challenging task due to the unstructured nature of SMS text data with non-linear complex correlations. In this concern, considering the recent advancements in the domain of cybersecurity, we have proposed a hybrid deep learning framework that extracts robust features from SMS texts followed by an automatic detection of Phishing SMS. Due to combining the potential capability of individual models into one hybrid framework, it has outperformed various other individual machine learning and deep learning models. The proposed Phishing Detection framework is an effective hybrid combination of pretrained transformer model, MPNet (Masked and Permuted Language Modeling), with supervised ConvNets (CNN) and Bi-directional Gated Recurrent Units (GRU). It is intended to successfully detect unstructured short phishing text messages that contain complex patterns.

As the echoes of ChatGPT's remarkable success continue to permeate the AI community, its formidable successor, GPT-4, has emerged, showing off a wealth of novel features. In this concise paper, we elucidate the capabilities of GPT-4 and conduct a comparative analysis with its predecessor, ChatGPT, offering insights into their relative strengths and advancements in the rapidly evolving field of generative AI. We also present a comprehensive summary of the major performance results reported by OpenAI on GPT-4 across various Natural Language Processing (NLP) tasks. We place great emphasis on the innovative advancements offered by GPT-4 in comparison to its predecessors. Our focus is on highlighting its remarkable performance while also mentioning its limitations. The purpose of this paper is to deliver a succinct understanding of the new features and performance benchmarks of GPT-4.

The keyless systems have replaced the old fashion methods of inserting physical keys in the keyhole to, i.e., unlock the door, because they are inconvenient and easy to be exploited by the threat actors. Keyless systems use the technology of radio frequency (RF) as an interface to transmit signals from the key fob to the vehicle. However, Keyless systems are susceptible to being compromised by a thread actor who intercepts the transmitted signal and performs a reply attack. In this paper, we propose a transfer learning-based model to identify the replay attacks launched against remote keyless controlled vehicles. Specifically, the system makes use of a pre-trained ResNet50 deep neural network to predict the wireless remote signals used to lock or unlock doors of a remote-controlled vehicle system remotely. The signals are finally classified into three classes: real signal, fake signal high gain, and fake signal low gain. We have trained our model with 100 epochs (3800 iterations) on a KeFRA 2022 dataset, a modern dataset. The model has recorded a final validation accuracy of 99.71% and a final validation loss of 0.29% at a low inferencing time of 50 ms for the model-based SGD solver. The experimental evaluation revealed the supremacy of the proposed model.

GPT-4 was released in March 2023 to wide acclaim, marking a very substantial improvement across the board over GPT-3.5 (OpenAI's previously best model, which had powered the initial release of ChatGPT). Despite the genuinely impressive improvement, however, there are good reasons to be highly skeptical of GPT-4's ability to reason. This position paper discusses the nature of reasoning; criticizes the current formulation of reasoning problems in the NLP community and the way in which the reasoning performance of LLMs is currently evaluated; introduces a collection of 21 diverse reasoning problems; and performs a detailed qualitative analysis of GPT-4's performance on these problems. Based on the results of this analysis, the paper argues that, despite the occasional flashes of analytical brilliance, GPT-4 at present is utterly incapable of reasoning.

The rapid proliferation of information in the digital era underscores the importance of accurate historical representation and interpretation. While artificial intelligence (AI) has shown promise in various fields, its potential for historical fact-checking and gap-filling remains largely untapped. This study evaluates the performance of three large language models (LLMs)—GPT-3.5, GPT-4, and Google- BARD—in the context of predicting and verifying historical events based on given data. A novel metric, "Distance to Reality" (DTR), is introduced to assess the models' outputs against established historical facts. The results reveal a substantial potential for AI in historical studies, with GPT-4 demonstrating superior performance. This paper underscores the need for further research into AI's role in enriching our understanding of the past and bridging historical knowledge gaps.

Generative artificial intelligence tools are flourishing across various sciences, particularly in the educational domain. ChatGPT, with its versions 3.5 and 4.0, stands out for its multilingual capabilities, catering to both students and educators. This research aimed to evaluate the performance differences between these versions, and importantly, to compare the accuracy of ChatGPT's responses in English versus Arabic. In our study, we presented ChatGPT with 39 chemistry problems related to 6th-7th grade curriculum, including twenty-five open questions and fourteen multiple-choice questions. Each response was categorized as accurate, partially accurate, or inaccurate. Our analysis, focused on trends across versions and languages, revealed significant enhancements in version 4.0, particularly in its ability to process Arabic. However, despite these improvements, our findings indicate that responses in English consistently outperformed those in Arabic in terms of accuracy. In light of this discrepancy, we suggest that ChatGPT either incorporates additional Arabic databases into its training regimen or develops a code that adopts the following strategy to ensure more accurate responses in science education: questions should first be translated from Arabic to English, addressed, and then retranslated into Arabic before being presented to the user. This approach takes advantage of the greater accuracy of responses in English to benefit Arabic-speaking users, and it may also improve outcomes for speakers of other languages who experience less precise answers in the fields of chemistry or science education.

One of the most rapidly advancing areas of deep learning research aims at creating models that learn to disentangle the latent factors of variation from a data distribution. However, modeling joint probability mass functions is usually prohibitive, which motivates the use of conditional models assuming that some information is given as input. In the domain of numerical cognition, deep learning architectures have successfully demonstrated that approximate numerosity representations can emerge in multi-layer networks that build latent representations of a set of images with a varying number of items. However, existing models have focused on tasks requiring to conditionally estimate numerosity information from a given image. Here we focus on a set of much more challenging tasks, which require to conditionally generate synthetic images containing a given number of items. We show that attention-based architectures operating at the pixel level can learn to produce well-formed images approximately containing a specific number of items, even when the target numerosity was not present in the training distribution.

Objective: Mechanical bowel preparation is essential for the visualisation of the colon during colonoscopy, and previous studies have identified changes in gut microbiota composition after bowel preparation and colonoscopy. Considering the gut microbiota is increasingly implicated in mood, we aimed to explore the potential impact of this intervention on mood and the microbiota-gut-brain axis. Methods: We conducted a pre-post intervention study, with timepoints of one week before and one month after bowel preparation and colonoscopy, in community-dwelling adults referred for colonoscopy. Our primary outcome was change in average Hospital Anxiety and Depression Scale depression sub-scores using generalised estimating equations. We also examined changes in average anxiety, stress, and quality of life scores, and faecal microbiota composition and potential function using 16S rRNA sequencing. We further explored associations between changes in depressive symptoms and the gut microbiota and conducted post-hoc analyses to explore potential effect modifiers. Key Results: 59 participants were included in the analyses. Depressive symptoms decreased one-month post-procedure compared to baseline (adjusted β=−0.64; 95%CI: −1.18, −0.11). Irritable bowel syndrome (IBS) appeared to moderate this relationship (1.78; 0.292, 3.26); depressive symptoms increased in those with, and decreased in those without, IBS. Changes in gut microbiome composition included reduced alpha diversity, modest effects on beta-diversity, and increases in genera associated with potential health benefits, one-month post-procedure compared to baseline. Increases in the CLR-transformed abundance of Ruminococcaceae UCG-009 were associated with improvements in depressive symptoms. Conclusions: There is preliminary evidence of a potential mental health effect of bowel preparation and colonoscopy, which may be associated with changes to the gut microbiota. There may also be a differential mental health impact of this procedure for those with and without IBS. Considering the limitations of our study, further research is required to confirm these findings and their clinical relevance.

The predominance of English and Latin-based large language models (LLMs) has led to a notable deficit in native Arabic LLMs. This discrepancy is accentuated by the prevalent inclusion of English tokens in existing Arabic models, detracting from their efficacy in processing native Arabic's intricate morphology and syntax. Consequently, there is a theoretical and practical imperative for developing LLMs predominantly focused on Arabic linguistic elements. To address this gap, this paper proposes ArabianGPT, a series of transformer-based models within the ArabianLLM suite designed explicitly for Arabic. These models, including ArabianGPT-0.1B and ArabianGPT-0.3B, vary in size and complexity, aligning with the nuanced linguistic characteristics of Arabic. The AraNizer tokenizer, integral to these models, addresses the unique morphological aspects of Arabic script, ensuring more accurate text processing. Empirical results from fine-tuning the models on tasks like sentiment analysis and summarization demonstrate significant improvements. For sentiment analysis, the fine-tuned ArabianGPT-0.1B model achieved a remarkable accuracy of 95%, a substantial increase from the base model's 56%. Similarly, in summarization tasks, fine-tuned models showed enhanced F1 scores, indicating improved precision and recall in generating concise summaries. Comparative analysis of fine-tuned ArabianGPT models against their base versions across various benchmarks reveals nuanced differences in performance, with fine-tuning positively impacting specific tasks like question answering and summarization. These findings underscore the efficacy of fine-tuning in aligning ArabianGPT models more closely with specific NLP tasks, highlighting the potential of tailored transformer architectures in advancing Arabic NLP.

Current neural network models can demonstrate reasonable-looking behavior, considered by some developers and researchers human-like. For example, a large language model GPT-3 is susceptible to human-like cognitive biases. Yet there is no data of such models solving emotional intelligence (EI) tasks. They are connected to the abilities that has been previously considered as specifically human EI is an important aspect of human communication. The ability to understand and respond to emotional cues is essential for effective communication. Therefore, it is crucial to determine the ways AI models such as ChatGPT demonstrate EI. The present research aims to measure the EI of GPT-4, a large language model trained by OpenAI. Russian version of the Mayer–Salovey–Caruzo Emotional Intelligence Test sections B, C, D, F, G and H were used in this research. High points were obtained in Understanding emotions scale and Strategic EI. Mean points are obtained in Managing emotions scale. Low and less reliable values are obtained in Using emotions to facilitate thought scale. Thus, GPT-4 seems already capable of identifying emotions in text and describing techniques for managing them. However, complex cases and irregular situations requiring emotions qualitative analysis would be a hard task for GPT-4.

Underwater image often exhibit detail blurring and color distortion due to light scattering, impurities, and other influences, obscuring essential textures and details. This presents a challenge for existing super-resolution techniques in identifying and extracting effective features, making high-quality reconstruction difficult. This research aims to innovate underwater image super-resolution technology to tackle this challenge. Initially, an underwater image degradation model was created by integrating random subsampling, Gaussian blur, mixed noise, and suspended particle simulation to generate a highly realistic synthetic dataset, thereby training the network to adapt to various degradation factors. Subsequently, to enhance the network's capability to extract key features, improvements were made based on the symmetrically structured Blind Super-Resolution Generative Adversarial Network (BSRGAN) model architecture. An attention mechanism based on energy functions was introduced within the generator to assess the importance of each pixel, and a weighted fusion strategy of adversarial loss, reconstruction loss, and perceptual loss was utilized to improve the quality of image reconstruction. Experimental results demonstrate that the proposed method achieved significant improvements in Peak Signal-to-Noise Ratio (PSNR) and Underwater Image Quality Measure (UIQM) by 0.85 dB and 0.19, respectively, significantly enhancing the visual perception quality and indicating its feasibility in super-resolution applications.

This article delves into the intersection of generative AI and digital twins within drug discovery, exploring their synergistic potential to revolutionize pharmaceutical research and development. Through various instances and examples, we illuminate how generative AI algorithms, capable of simulating vast chemical spaces and predicting molecular properties, are increasingly integrated with digital twins of biological systems to expedite drug discovery. By harnessing the power of computational models and machine learning, researchers can design novel compounds tailored to specific targets, optimize drug candidates, and simulate their behavior within virtual biological environments. This paradigm shift offers unprecedented opportunities for accelerating drug development, reducing costs, and, ultimately, improving patient outcomes. As we navigate this rapidly evolving landscape, collaboration between interdisciplinary teams and continued innovation will be paramount in realizing the promise of generative AI and digital twins in advancing drug discovery.

This paper examines the potential of Artificial Intelligence (AI) to address societal megatrends, with a specific focus on OpenAI’s Generative Pre-Trained Transformer 3 (GPT-3). To do this, we con-ducted an analysis using GPT-3 in order to explore the benefits of AI for digitalization, urbanization, globalization, climate change, automation and mobility, global health issues and aging population. We also looked at emerging markets as well as sustainability in this study. Interaction with GPT-3 was done solely through prompt questions and generated responses were analyzed. Our results indicate that AI can significantly improve our understanding of these trends by providing insights into how they develop over time and what solutions could be implemented. Further research is needed to determine how effective AI will be in addressing them successfully but initial findings are encouraging. Our discussion focuses on the implications of our findings for society going forward and suggests that further investigation should be done into how best to utilize new technologies like GPT-3 when tackling these challenges. Lastly, we conclude that while there is still much work left to do before any tangible effects can be seen from utilizing AI tools such as GPT-3 on societal mega-trends; early indications suggest it may have a positive impact if used correctly.

Previous camera self-calibration methods have exhibited certain notable shortcomings. On one hand, they either exclusively emphasized scene cues or solely focused on vehicle-related cues, resulting in a lack of adaptability to diverse scenarios and a limited number of effective features. Furthermore, these methods either solely utilized geometric features within traffic scenes or exclusively extracted semantic information, failing to comprehensively consider both aspects. This limited the comprehensive feature extraction from scenes, ultimately leading to a decrease in calibration accuracy. Additionally, conventional vanishing point-based self-calibration methods often required the design of additional edge-background models and manual parameter tuning, thereby increasing operational complexity and the potential for errors. Given these observed limitations, and in order to address these challenges, we propose an innovative roadside camera self-calibration model based on the Transformer architecture. This model possesses a unique capability to simultaneously learn scene features and vehicle features within traffic scenarios while considering both geometric and semantic information. Through this approach, our model can overcome the constraints of prior methods, enhancing calibration accuracy and robustness while reducing operational complexity and the potential for errors. Our method outperforms existing approaches on both real-world dataset scenarios and publicly available datasets, demonstrating the effectiveness of our approach.

Ultrasound (US) has become a widely used imaging modality in clinical practice, characterized by its rapidly evolving technology, advantages, and unique challenges such as low imaging quality and high variability. There is a critical need to develop advanced automatic US image analysis methods to enhance diagnostic accuracy and objectivity. Vision transformer, a recent innovation in machine learning, has demonstrated significant potential in various research fields, including general image analysis and computer vision, due to its capacity to process large datasets and learn complex patterns. Its suitability for automatic US image analysis tasks, such as classification, detection, and segmentation, has been recognized. This review provides an introduction to vision transformer and discusses its applications in specific US image analysis tasks, while also addressing the open challenges and potential future trends in its application in medical US image analysis. Vision transformer has shown promise in enhancing the accuracy and efficiency of ultrasound image analysis and is expected to play an increasingly important role in the diagnosis and treatment of medical conditions using ultrasound imaging as technology progresses.

Transformers are increasing replacing older generation of deep neural networks due to their success in a wide range of application. The dominant approach of using transformers is to pre-train them on a large training dataset and then fine-tune them on a downstream task. However, as transformers becoming larger, the fine-tuning approach is become an infeasible approach for transfer learning. In this short survey, we list a few recent methods that makes using transformers based on transfer learning more efficient.

In power systems there are complex transformer structures, whose accurate analysis is not possible using the techniques available today. This paper presents a systematic data driven analysis method for coupled inductors of arbitrary complexity. The method first establishes a winding matrix N mapping the windings to the limbs of the transformer. A permeance matrix P is created from the reluctance network of the magnetic core. A generalized inductance matrix L mapping currents in the transformer windings to the induced voltages is generated based on the winding (N) and permeance (P) matrices. The inductance matrix representation of a coupled inductor is then transformed to an admittance matrix, which can be integrated to the nodal analysis of the electrical circuit surrounding the coupled inductor. The method presented is validated by simulations with real transformer structures using electromagnetic transient program (EMTP/ATP).

According to many researchers, Abstraction is the basis of mathematics, computing, counting devices, and computer science and engineering. What is more, all of the above deal with complexity management in some way, and abstraction is the most basic mechanism of complexity management.Generative software development - whether in the sense of empowering humans by machine to create software or in the sense of reusing products - has been and is one of the serious concerns and goals of software engineering. The interesting thing is that in both views of generativity, the main issue is still, in a way, complexity management: whether this complexity management is to achieve diversity and reuse management (Czarnecki’s approach) or to Structuring from existing structures (the approach of Alexander and his followers in an object-oriented society).In this article, we will first look at complexity and its various definitions. The definitions that show, despite the different perspectives on complexity in different disciplines and domains, all point in one direction. We will conclude that complexity is rooted in multiplicity. In the following, we will formally define complexity. In the following discussion, we will look at the generative patterns of software development, and then we will look at the complexity management patterns at seven levels.In this article, the author has tried to maintain a comprehensive approach to complexity and to consider the approaches of different domains to complexity.

In the realm of computer vision, semantic segmentation is the task of recognizing objects in images at the pixel level. This is done by performing a classification of each pixel. The task is complex and requires sophisticated skills and knowledge about the context to identify objects’ boundaries. The importance of semantic segmentation in many domains is undisputed. In medical diagnostics, it simplifies the early detection of pathologies, thus mitigating the possible consequences. In this work, we provide a review of the literature on deep ensemble learning models for polyp segmentation and we develop new ensembles based on convolutional neural networks and transformers. The development of an effective ensemble entails ensuring diversity between its components. To this end, we combine different models (HarDNet-MSEG, Polyp-PVT, and HSNet) trained with different data augmentation techniques, optimization methods, and learning rates, which we experimentally demonstrate to be useful to form a better ensemble. Most importantly, we introduce a new method to obtain the segmentation mask which is more suitable for combining transformers in an ensemble. In our extensive experimental evaluation, the proposed ensembles exhibit state-of-the-art performance.

Image classification in Remote Sensing and Geographic Information Systems (GIS) containing various land-cover classes is essential for efficient and sustainable land-use estimation, and other tasks like object detection, localization and segmentation. Deep Learning (DL) techniques have shown a tremendous potential in the GIS domain. While Convolutional Neural Networks (CNN) have dominated most of the image analysis domain, a new architecture called transformers have proved to be a unifying solution for several AI-based processing pipelines. Vision Transformers (ViT), a variant of transformers can have comparable and in some cases better accuracy than a CNN. However, they suffer from a significant drawback associated with an excessive use of training parameters. In this research we explore several modifications in the vision transformer architectures, especially MobileViT that can be optimized while boosting accuracy. To verify our proposed approach these new architectures are trained on four land-cover datasets AID, EuroSAT, UC-Merced, and WHU-RS19. Experiments reveal that combination of lightweight convolutional layers including ShuffleNet along with depthwise separable convolutions and average pooling can reduce the trainable parameters by 17.85% and yet achieve higher accuracy than the base MobileViT. It is also observed that utilizing a combination of convolution layers along with multi-headed self attention layers in MobileViT variants provide better performance in capturing local and global features unlike the standalone ViT architecture that utilizes almost 95% more parameters than the proposed MobileViT variant.

In this work we present the results obtained with an artificial neural network (ANN) which we trained to predict the expected output of high-order harmonic generation (HHG) process, while exploring a multi-dimensional parameter space. We argue on the utility and efficiency of the ANN model and demonstrate its ability to predict the outcome of HHG simulations. In this case study we present the results for a loose focusing HHG beamline, where the changing parameters are: the laser pulse energy, gas pressure, gas cell position relative to focus and gas cell length. The physical quantity which we predict here using ANN is directly related to the total harmonic yield in a specified spectral domain (20-40 eV). We discuss the versatility and adaptability of the presented method.

Artificial intelligence (AI) is a current trend in computer science, which extends itself its amazing capacities to other technologies such as mechatronics and robotics. Going beyond technological applications, the philosophy behind AI is that there is a vague and potential convergence of artificial manufacture and natural world although the limiting approach may be still very far away, but why? The implicit problem is that Darwin theory of evolution focuses on natural world where breeding conservation is the cornerstone of the existence of creature world but there is no similar concept of breeding conservation in artificial world whose things are created by human. However, after developing for a long time until now, AI issues an interesting concept of generation in which artifacts created by computer science can derive their new generations inheriting their aspects / characteristics. Such generated artifacts make us look back on offsprings by the process of breeding conservation in natural world. Therefore, it is possible to think that AI generation, which is a recent subject of AI, is a significant development in computer science as well as high-tech domain. AI generation does not help us to reach near biological evolution even in the case that AI can combine with biological technology but, AI generation can help us to extend our viewpoint about Darwin theory of evolution as well as there may exist some uncertain relationship between man-made world and natural world. Anyhow AI generation is a current important subject in AI and there are two main generative models in computer science: 1) generative model that applies large language model into generating natural language texts understandable by human and 2) generative model that applies deep neural network into generating digital content such as sound, image, and video. This technical report focuses on deep generative model (DGM) for digital content generation, which is a short summary of approaches to implement DGMs. Researchers can read this work as an introduction to DGM with easily understandable explanations.

Background: Several studies recently confirmed the emergence of resistance to antimalarial drugs in sub-Saharan Africa. Multiple first-line treatment (MFT) is one of the measures envisaged to respond to the emergence and spread of this resistance. The aim of this study was to identify perceived advantages and disadvantages of several MFT deployment strategies and to better understand potential implementation drivers and barriers. Methods: A qualitative survey was conducted in seven sub-Saharan countries amongst key opinion-leaders, national decision-makers, and end users. A total of 200 individual interviews were conducted and findings were analyzed following a thematic inductive approach. Results: From a policy perspective, the new MFT intervention would require endorsement at the global, national, and regional levels to ensure its inclusion in guidelines. Funding of the MFT intervention could be a bottleneck due to costs associated with additional training of healthcare workers, adaptation of drug delivery mechanisms, and higher costs of drugs. Concerning the MFT deployment strategies, a slight preference for the segmentation strategy was expressed over the rotation and geographic approaches, due to the perception that a segmentation approach is already in place at country level. Conclusions: The findings highlighted the need for a collective approach to MFT deployment through the engagement of stakeholders at all levels of malaria management.

This paper applies a new artificial intelligence architecture, the Temporal Fusion Transformer (TFT), for the joint GDP forecasting of 25 OECD countries at different time horizons. This new attention-based architecture offers significant advantages over other deep learning methods. First, results are interpretable since the impact of each explanatory variable on each forecast can be calculated. Second, it allows to visualize persistent temporal patterns and to identify significant events and different regimes. Third, it provides quantile regressions and permits to train the model on multiple time series from different distributions. Results suggest that TFTs outperform regression models, especially in periods of turbulence such as the COVID-19 shock. Interesting economic interpretations are obtained depending on whether the country is domestic demand-led or export-led growth. In essence, TFT is revealed as a new tool that artificial intelligence provides to economists and policy makers, with enormous prospects for the future.

Over the past decade, research in the field of Deep Learning has brought about novel improvements in image generation and feature learning; one such example being a Generative Adversarial Network. However, these improvements have been coupled with an increasing demand on mathematical literacy and previous knowledge in the field. Therefore, in this literature review, I seek to introduce Generative Adversarial Networks (GANs) to a broader audience by explaining their background and intuition at a more foundational level. I begin by discussing the mathematical background of this architecture, specifically topics in linear algebra and probability theory. I then proceed to introduce GANs in a more theoretical framework, along with some of the literature on GANs, including their architectural improvements and image-generation capabilities. Finally, I cover state-of-the-art image generation through style-based methods, as well as their implications on society.

Unmanned Aerial Vehicles (UAVs) equipped with hyperspectral imagers have emerged as an essential technology for the characterization of inland water bodies. The high spectral and spatial resolutions of these systems enable the retrieval of a plethora of optically-active water quality parameters via band ratio algorithms and machine learning methods. However, fitting and validating these models requires access to sufficient quantities of in situ reference data which are time-consuming and expensive to obtain. In this study, we demonstrate how the Generative Topographic Mapping (GTM), a Bayesian realization of the Self-organizing Map, can be used to visualize high-dimensional hyperspectral imagery and extract spectral signatures corresponding to unique endmembers present in the water. Using data collected across a North Texas pond, we first apply the GTM to visualize the distribution of captured reflectance spectra revealing small-scale spatial variability of water composition. Next, we demonstrate how the nodes of the fitted GTM can be interpreted as unique spectral endmembers. Using extracted endmembers together with the normalized spectral similarity score, we are able to efficiently map the abundance of near shore algae as well as the evolution of a rhodamine tracer dye used to simulate water contamination by a localized source.

The COVID-19 pandemic exposed the vulnerability of current vaccine technologies characterized by a slow onset of action and antigen-specific immune response. Although parental vaccines offer long-term protection against homologous strains, they rely exclusively on adaptive immune memory to produce neutralizing antibodies that are ineffective against new vaccine variants. Moreover, growing evidence highlights the multifaceted functions of trained immunity to elicit a rapid and enhanced innate response against unrelated stimuli or pathogens to subsequent triggers. This review discusses the protective role of trained immunity against respiratory pathogens and the experimental models essential for evaluating novel inducers of trained immunity. We further elaborate on the potential of trained immunity to leverage protection against emerging pathogens via recognition of diverse antigens by pathogen recognition receptors (PPRs) on innate immune cells. We also propose integrating trained- with adaptive- immunity to shape next-generation vaccines by coupling each one's unique characteristics.

The study introduces a novel method to identify potential citation manipulation in academic papers using perturbations of a deep embedding model, incorporating Graph Masked Autoencoders. This approach integrates textual information with graph connectivity evidence, resulting in a more sophisticated model of citation distribution. By training a deep network using partial data and reconstructing masked connections, the method leverages the inherent characteristics of central connections under network perturbations. Quantitative evaluations demonstrate its remarkable ability to pinpoint trustworthy citations in the analyzed data and raise concerns about potentially unreliable references due to potential manipulation.

This study explored the development of neuromuscular fatigue responses during progressive cycling exercise. The sample comprised 32 participants aged 22.0 ± 0.54 years who were assigned into three groups: endurance-trained group (END, triathletes, n = 10), strength-trained group (STR, body-builders, n = 10), and control group (CON, recreationally active students, n = 12). The incremental cycling exercise was performed using a progressive protocol starting with a 3 min resting measurement and then 50 W workload with subsequent constant increments of 50 W every 3 minutes until 200 W. Surface electromyography (SEMG) of rectus femoris muscles was recorded during the final 30 s of each of the four workloads. During the final 15 s of each workload, participants rated their overall perception of effort using the 20-point RPE scale. Post-hoc Tukey’s HSD testing showed significant differences between END and STR groups in MF and MPF across all workloads (p < 0.001 and p < 0.01, respectively). Athletes from END group had significantly lower AEMG responses than those from STR (p = 0.0093) and CON group (p = 0.0006). Increasing RPE points from 50 to 200 W were significantly higher in STR than in END group (p < 0.001). In conclusion, there is a significant variation in neuromuscular fatigue profiles between athletes with different training backgrounds when a cycling exercise is applied. The approximately linear trend of the SEMG and RPE values of both groups of athletes with increasing workload support the increased skeletal muscle recruitment with perceived exertion or fatiguing effect.

Kinetic modeling is an essential tool in systems biology research, enabling the quantitative analysis of biological systems and predicting their behavior. However, the development of kinetic models is a complex and time-consuming process. In this article, we propose a novel approach called KinModGPT, which generates kinetic models directly from natural language text. KinModGPT employs GPT-3 as a natural language interpreter and Tellurium as an SBML generator. We demonstrate the effectiveness of KinModGPT in creating SBML models from complex natural language descriptions of biochemical reactions. KinModGPT successfully generates valid SBML models from a range of natural language model descriptions of metabolic pathways, protein-protein interaction networks, and heat shock response. This article demonstrates the potential of KinModGPT in kinetic modeling automation.

Artificial Intelligence (AI) experienced significant advancements in recent years, and its impact is already recognized across various industries. Yet, the rise of AI has led to a growing concern about its impact on meeting the Sustainable Development Goals (SDGs). The aim of this paper was to evaluate contributions and the potential impact of AI to sustainable development in the society domain. Furthermore, the study analyzed GPT-3 responses, as one of the largest language models developed by OpenAI, descriptively. We conducted a set of queries on the SDGs to gather infor-mation on GPT-3`s perceptions of AI impact on sustainable development. Analysis of GPT-3’s contribution potential towards the SDGs showcased its broad range of capabilities for contributing to the SDGs in areas such as education, health, and communication. The study findings provide valuable insights into the contributions of AI to sustainable development in the society domain and highlight the importance of proper regulations to promote the responsible use of AI for sustainable development. We plan subsequent research for effects on ecologic and economic domains of SDGs. We highlighted the potential for improvement in neural language processing skills of GPT-3 by avoiding imitating weak human writing styles with more mistakes in longer texts.

Person Re-identification is the task of recognizing comparable subjects across a network of nonoverlapping cameras. This is typically achieved by extracting from the source image a vector of characteristic features of the specific person captured by the camera. Learning a good set of robust, invariant and discriminative features is a complex task, often leveraging contrastive learning. In this article, we explore a different approach, learning the representation of an individual as the conditioning information required to generate images of the specific person starting from random noise. In this way we decouple the identity of the individual from any other information relative to a specific instance (pose, background, etc.), allowing interesting transformations from one identity to another. As generative models, we use the recent diffusion models that have already proven their sensibility to conditioning in many different contexts. The results presented in this article serve as a proof-of-concept. While our current performance on common benchmarks is lower than state-of-the-art techniques, the approach is intriguing and rich of innovative insights, suggesting a wide range of potential improvements along various lines of investigation.

Generative artificial intelligence (GenAI) has been developing with many incredible achievements like ChatGPT and Bard. Deep generative model (DGM) is a branch of GenAI, which is preeminent in generating raster data such as image and sound due to strong points of deep neural network (DNN) in inference and recognition. The built-in inference mechanism of DNN, which simulates and aims to synaptic plasticity of human neuron network, fosters generation ability of DGM which produces surprised results with support of statistical flexibility. Two popular approaches in DGM are Variational Autoencoders (VAE) and Generative Adversarial Network (GAN). Both VAE and GAN have their own strong points although they share and imply underline theory of statistics as well as incredible complex via hidden layers of DNN when DNN becomes effective encoding/decoding functions without concrete specifications. In this research, VAE and GAN is unified into a consistent and consolidated model called Adversarial Variational Autoencoders (AVA) in which VAE and GAN complement each other, for instance, VAE is good at generator by encoding data via excellent ideology of Kullback-Leibler divergence and GAN is a significantly important method to assess reliability of data which is realistic or fake. In other words, AVA aims to improve accuracy of generative models, besides AVA extends function of simple generative models. In methodology this research focuses on combination of applied mathematical concepts and skillful techniques of computer programming in order to implement and solve complicated problems as simply as possible.

Due to the long-term deformation settlement of foundations, issues such as damage and functional failure of buildings and structures have long been of concern in the engineering field. The creep of soil is one of the primary causes leading to long-term deformation of foundations. In this paper, the consolidation deformation, creep characteristics, and creep model of reconstituted saturated silty clay were studied by isotropic compression test and triaxial shear creep test. The results show that for the isotropic compression test, although the applied load adopts different stages of loading, as long as the final applied confining pressure is the same, the number of stages applied by the confining pressure has little effect on the final isotropic consolidation deformation of the sample and the triaxial undrained shear strength after creep. However, for the triaxial shear creep test, it is found that under the same final deviatoric stress, the final deviatoric strain of the sample is closely related to the number of loading stages of deviatoric stress. It shows that the more loading stages of the sample loaded with the same deviatoric stress, the smaller the final deviatoric strain, and the triaxial undrained shear strength of the sample after creep will increase. In addition, it is reasonable to determine the pore pressure dissipation of the sample to 95%((u0-u0)/u0=95%) as the main consolidation end time (tEOP) of the sample. The isotropic consolidation creep curves and the triaxial shear creep curves show certain nonlinearity. Then the logarithmic model and the hyperbolic model were used to fit the creep curves of the samples. It is found that the hyperbolic model has a better fitting effect than the logarithmic model, but for the triaxial shear creep test, the creep parameters of the sample change greatly. Therefore, studying the creep characteristics of soil under different pre-loading steps is of significant engineering importance for evaluating the long-term deformation of underground structures.

Construction work is physically demanding. The tasks involved in this professional occupation physically exhaust workers and put them at high risk of injury. This work aims to analyze muscle fatigue in civil construction workers, with and without physical activity historical practice. For this purpose, the muscle activity of the Biceps Femoris, Trapezius Transversalis, Erector Spinae Longissimus, was evaluated using surface electromyography. Eleven male civil construction workers aged 26 to 52 years (38.00 ± 9.60), of whom 7 were untrained (N =7) and 4 were trained (N=4) participated in this study. Each subject completed two assessments at two different times. At the first assessment moment, a questionnaire was completed, and the maximum repetition of the deadlift exercise was assessed. At the second assessment, the subjects were assessed while performing the deadlift in four different situations: 50%RM at rest, 80%RM at rest, 80%RM at fatigue, and 50%RM at fatigue. The trained construction workers have higher levels of muscle activity, and there are no significant differences in muscle activation between the fatigued and non-fatigued sets. There appears to be a strong influence of physical exercise on muscle activity and fatigue in civil construction workers.

Real-time tracking is one of the most challenging problems in computer vision. Most Transformer-based trackers usually require expensive computational and storage power, which leads to the inability of these robust trackers to achieve real-time performance in resource-constrained devices. In this work, we propose a lightweight tracker, AnteaTrack. To localize the target more precisely, we develop a scaling-invariant max-filtering operator employing a sliding window combined with local max-pooling, which filters out the suspected target from the feature and performs an augmented representation while suppressing the background. For a more tight bounding box, we employ Pixel-Shuffle to increase the resolution of the feature map and get a more fine-grained representation of the target. In addition, AnteaTrack can run in real-time at 47 frames per second(FPS) on the CPU. We tested AnteaTrack on 5 datasets, and a large number of experiments have shown that AnteaTrack provides the most efficient solution compared to the same type of CPU real-time trackers. The code will be available at https://github.com/cnchange/AnteaTrack.

The most common source of transformer failure is in the insulation, and the most prevalent warning signal for insulation weakness is partial discharge (PD). Locating positons of these partial discharges would help repair the transformer to prevent failures. This work investigates algorithms that could be deployed to locate the position of a PD event using data from ultra-high frequency (UHF) sensors inside the transformer. These algorithms typically proceed in two steps: first determine the signal arrival time and then locate the position based on time differences. This paper reviews available methods for each task and then propose new algorithms: a convolutional iterative filter with thresholding (CIFT) to determine the signal arrival time and a reference table of travel times to resolve the source location. The effectiveness of these algorithms are tested with a set of laboratory-triggered PD events and two sets of simulated PD events inside transfers in production use. Tests show the new approach provides more accurate locations than the best-known data analysis algorithms, and the difference is particularly large, 3.7X, when the signal sources are far from sensors.

Deep convolutional neural networks (CNNs) have been widely used in various medical imaging tasks. However, due to the intrinsic locality of convolution operation, CNNs generally cannot model long-range dependencies well, which are important for accurately identifying breast cancer from unregistered multi-view mammograms. This motivates us to leverage the architecture of Vision Transformers to capture long-range relationships of multiple mammograms from the same patient. For this purpose, we employ local Transformer blocks to separately learn patch relationships within a specific view (CC/MLO) of one side (right/left) of mammogram. The outputs from different views and sides are concatenated and fed into global Transformer blocks, to jointly learn patch relationships between two different views of the left and right breasts. To evaluate the proposed model, we retrospectively assembled a dataset involving 949 sets of CC and MLO view mammograms, which include 470 malignant cases and 479 normal or benign cases. We trained and evaluated the model using a five-fold cross-validation method. Without any arduous preprocessing steps (e.g., optimal window cropping, chest wall or pectoral muscle removal, two-view image registration, etc.), our two-view Transformer-based model achieves lesion classification accuracy of 77.0% and area under ROC curve (AUC = 0.814), outperforming state-of-the-art multi-view CNNs by 3.1% and 3%, respectively. Meanwhile, the new two-view model improves mammographic case classification accuracies of two single-view models by 7.4% (CC) and 4.5% (MLO), respectively. The promising results unveil the great potential of using Transformers to develop high-performing computer-aided diagnosis (CADx) schemes of mammography.

Artificial intelligence (AI) has the potential to revolutionize research by automating data analysis, generating new insights, and supporting the discovery of new knowledge. The top 10 contribution areas of AI towards public health were gathered in this feasibility study. We utilized the “text-davinci-003” model of GPT-3, using OpenAI playground default parameters. The model was trained with the largest training dataset any AI had, limited to a cut-off date in 2021. This study aimed to test the ability of GPT-3 to advance public health and to explore the feasibility of using AI as scientific co-author. The AI was asked for input including scientific quotations and the human authors reviewed responses for plausibility. We found that GPT-3 was able to assemble, summarize, and generate plausible text blocks relevant for public health concerns, elucidating valuable areas of application for itself. However, most quotations were invented by GPT-3 and thus, invalid. Ac-cording to today’s rules, we conclude that AI can contribute to public health research as a team member. Nevertheless, good scientific practice needs to be also followed for AI contributions, and a broad scientific discourse on AI contributions is needed. Policies for good scientific practice should be updated timely following this discourse.

An important indicator of human-related pollution in watersheds is dissolved oxygen (DO). The DO is highly dependent on both space and time characteristics of the watershed and is directly linked to eutrophication, which impairs the development of both the aquatic fauna and flora, also negatively impacting the water quality. Aspiring to reach a more accurate and precise forecasting approach to predict levels of DO, the present work proposes new graph-based and transformer-based deep learning models. The models were trained and validated for the Credit River Watershed, and the results were compared with both benchmarking and literature-found approaches. The proposed Graph Neural Network Sample and Aggregate (GNN-SAGE) model was the best-performing approach, reaching coefficient of determination (R2) and Root Mean Squared Error (RMSE) values of 97% and 0.34 ppm, respectively. The findings from the Shapley additive explanations (SHAP) indicated that the GNN-SAGE benefited from spatiotemporal information from the surrounding stations, improving the model’s results, and that temperature is a major input attribute for determining future DO levels. The results established that the proposed GNN-SAGE model stands as a state-of-the-art solution for DO forecasting, with potential for real-time water quality applications.

Chatbots are computer programs that use artificial intelligence to imitate human conversations. Recent advancements in deep learning have shown interest in utilizing language transformers, which do not rely on predefined rules and responses like traditional chatbots. This study provides a comprehensive review of previous research on chatbots that employ deep learning and transfer learning models. Specifically, it examines the current trends in using language transformers with transfer learning techniques to evaluate the ability of Arabic chatbots to understand conversation context and demonstrate natural behavior. The proposed methods explore the use of AraBERT, CAMeLBERT, AraElectra-SQuAD, and AraElectra (Generator/Discriminator) transformers, with different variants of these transformers and semantic embedding models. Two datasets were used for evaluation: one with 398 questions and corresponding documents, and another with 1395 questions and 365,568 documents sourced from Arabic Wikipedia. Extensive experimental works were conducted, evaluating both manually crafted questions and the entire set of questions, using confidence and similarity metrics. The experimental results showed that the AraElectra-SQuAD model achieved an average confidence score of 0.6422 and an average similarity score of 0.9773 on the first dataset, and an average confidence score of 0.6658 and similarity score of 0.9660 on the second dataset. The study concludes that the AraElectra-SQuAD model consistently outperformed other models, displaying remarkable performance, high confidence, and similarity scores, as well as robustness, highlighting its potential for practical applications in natural language processing tasks for Arabic chatbots. The study suggests that the AraElectra-SQuAD model can be further enhanced and applied in various tasks such as chatbots, virtual assistants, and information retrieval systems for Arabic-speaking users. By combining the power of transformer architecture with fine-tuning on SQuAD-like large data, this trend demonstrates its ability to provide accurate and contextually relevant answers to questions in Arabic.

Large power transformers are generally associated with a maximum capacity rating of 100 MVA or higher. These large liquid dielectric power transformers are a custom-built piece of equipment, thus very expensive, and a backbone element of the power grid. In extreme cases as, for example, severe geomagnetic disturbances, permanently monitoring their condition will enhance their electrical reliability and resilience to guarantee efficient management of its life cycle. However, some traditional monitoring/diagnosis techniques have singular features when applied to large power transformers and their interlinked subsystems. In this context, and since that information is hardly put in evidence and compiled in the literature, this paper reviews the particularities of monitoring and diagnosing those assets.

Traditional image steganography conceals secret messages into unprocessed natural images by modifying the pixel value, causing the obtained stego different from the original image in terms of statistical distribution, thereby could be detected by a well-trained classifier for steganalysis. To ensure the steganography is imperceptible and in line with the trend of art images produced by Artificial Intelligence Generated Content (AIGC) becoming popular in social networks, this paper proposes to embed hidden information throughout the process of the generation of an art-style image by designing an image style transformation neural network with steganography function. The proposed scheme takes a content image, an art-style image, and messages to be embedded as inputs, processing them with an encoder-decoder model, and finally generates a styled image containing the secret messages at the same time. An adversarial training technique is applied to enhance the imperceptibility of the generated art-styled stego image with from plain style-transferred images. The lack of the original cover image makes it difficult for the opponent learning a steganalyzer to identify the stego. The recommended approach can successfully withstand existing steganalysis techniques and attain the embedding capacity of 3 bits per pixel for a color image, according to experimental results.

In 2023, the technological landscape witnessed an unprecedented surge of transformative innovations, including Web 3.0 and machine learning, poised to revolutionize various aspects of our daily lives. At the forefront of this revolution stands artificial intelligence (AI), which has evolved from a concept once confined to science fiction into an indispensable force reshaping industries on a global scale. Notably, recent years have seen remarkable strides in AI functionalities, particularly in machine learning, computer vision, and natural language processing. These advancements have empowered AI to generate diverse interactions and media autonomously. This study ultimately provides invaluable insights into the dynamic intersection of AI and Advertising, con-tributing to a deeper comprehension of this evolving field. This study conducts a comprehensive review of Artificial Intelligence within the advertising domain, employing network analysis. By meticulously examining the evolution of AI through co-occurrences of terms, keywords, and co-authorship analysis, we aim to unveil the relationship of AI & Advertising through the lens of Generative models to identify pivotal trends, core concepts, and seminal studies that have profoundly influenced the industry. The analysis charts the trajectory leading to the emergence of Generative Artificial Intelligence (GenAI), a transformative development poised to revolutionize advertising practices and consumer engagement. In tandem with these objectives, the results highlight vital trends, emphasizing the growing prominence machine learning tools and techniques such as deep learning techniques and advanced natural language processing methods like word2vec, GANs, and more in shaping the future of advertising practices.

This survey proposed an evaluation model to analyze and examine different approaches to generativity. In addition to problem domain concepts, the following concepts were used to define this model: Complexity and complexity management, and Systematics view to achieve unified and integrated insight into disparate evaluation criteria. The research's approach to the said concepts is first introduced. Then, the evaluation model is presented.

We study the brain segmentation by dividing the brain into multiple tissues. Given possible brain segmentation by deep, machine learning can be efficiently exploited to expedite the segmentation process in the clinical practice. To accomplish segmentation process, a MRI and tissues transfer using generative adversarial networks is proposed. Given the better result, we propose the transfer model using GAN. For the case of the brain tissues, white matter (WM), gray matter (GM) and cerebrospinal fluid (CSF) are segmented. Empirical results show that this proposed model significantly improved segmentation results compared to the stat-of-the-art results. Furthermore, a dice coefficient (DC) metric is used to evaluate the model performance.

Mobile robots need an environmental perception ability in order to interact with the surrounding environment. In this paper we present the G-SLAM method, where the map is consisted of a cloud of scattered points in the continuous space and each point is accompanied by an obstacle existence probability. On the other hand the robot's pose (and trajectory) is estimated by an particle filters while the cloud of points is estimated by an adaptive recursive algorithm which is presented. Main feature of the presented method is the adaptive repositioning of the scattered points and their convergence around obstacles. In addition to our previous publications, In this paper a mathematical derivation of the method is presented and real case scenarios are discussed, presented and compared with other methods.

Abstract: The advent of Artificial Intelligence (AI) has in just a few years invested multiple areas of knowledge, also affecting the medical-scientific sector. An increasing number of AI-based applications have been developed, among which conversational AI has emerged. Among these, ChatGPT has risen to the headlines, scientific and otherwise, for its distinct propensity to simulate a 'real' discussion with its interlocutor, based on appropriate prompts. Although several clinical studies using ChatGPT have already been published in the literature, very little has yet been written about its potential application in human pathology. We conduct a systematic review following the Preferred Reporting Items for Systematic Re-views and Meta-Analyses (PRISMA) guidelines, using PubMed and Scopus as databases, with the fol-lowing keywords: ChatGPT OR Chat GPT, in combination with each of the following: Pathology, di-ag-nostic pathology, anatomic pathology. A total of 90 records were initially identified in the literature search, of which 6 were duplicates. After screening for eligibility and inclusion criteria, only 5 publications were ultimately included. The majority of publications were original articles (n = 2), followed by case reports (n = 1), letter to the editor (n = 1) and review (n = 1). Although the premises are exciting and ChatGPT is able to co-advise the pathologist in providing large amounts of scientific data for use in routine microscopic diagnostic practice, there are many limitations that need to be addressed and resolved, with the caveat that an AI-driven system should always provide support and never a decision-making motive during the anatomo-pathological diagnostic process.

Solid-state transformer (SST) is a promising technology for smart grids due to its flexible power control (better reliability) and high efficacy (by decreasing losses) compared with traditional transformers. The SST is designed in tandem with three stages, i.e., the input, isolation, and output stages. As three converter stages are used to design the SST to make a more reliable device in practice, it is highly prone to instabilities. Moreover, the key objective of SST design is to implement a modern power distribution system to make it more intelligent and reliable. The stability issue can be even more complicated by the presence of distributed energy resources in the distribution system. Thus, the stability of SST must be measured prior to /during the design, and to determine, a state-space model is used, which is developed in this paper. Each stage of SST is modeled with its controller, and the system's stability is measured through the controllability and observability test. Further, the conversion from state-space to transfer function model is done, and the stability of SST is shown using frequency and time-domain diagrams. With these, the different plots, namely, the Bode plot, Nyquist plot, Nichols chart, Root locus, pole-zero plot, and Eigen plot, are used to ascertain the relative and absolute stability. Finally, the SST Simulink model is tested and validated through hardwate-in-the-loop simulation, i.e., OPALRT, to show its effectiveness and applicability.

A key component for the proper functioning, availability, and reliability of power grids is the power trans- former. Although these are very reliable assets, the early detection of incipient degradation mechanisms are very important to prevent failures that may shorten their lifetime. In this work a review and comparative analysis, of classical Machine Learning algorithms (such as single and ensemble classification algorithms) and two automatic machine learning classifiers, are presented for the fault diagnosis of power transformers. The goal is to determine whether fully automated ML approaches are worse or better than traditional ML frameworks that require a human in the loop (such as a data scientist) to identify transformer faults from oil-dissolved gases data. The methodology uses a DB compiled from published and proprietary transformer fault samples. Faults data is obtained from the literature, the Duval pentagon method, and user-expert knowledge. The parameters from, either single or ensemble classifiers, were optimized through standard machine learning procedures. The results showed that the best performing algorithm is a robust automatic machine learning classifiers model followed by classical algorithms such as neural networks and stacking ensembles. These results highlight the ability of a robust automatic machine learning model for, handling imbalanced power transformers faults datasets with high accuracy, using the minimum tuning effort by the electrical experts. By providing the most probable transformer fault condition will reduce the time required to find and solve the fault.

Since large power transformers are custom-made, and their design process is a labor-intensive task, their design process is split into different parts. In tendering, the price calculation is based on the preliminary design of the transformer. Due to the complexity of this task, it belongs to the most general branch of discrete, non-linear mathematical optimization problems. Most of the published algorithms are using a copper filling factor based winding model to calculate the main dimensions of the transformer during this first, preliminary design step. Therefore, these cost optimization methods are not considering the detailed winding layout and the conductor dimensions. However, the knowledge of the exact conductor dimensions is essential to calculate the thermal behaviour of the windings and make a more accurate stray loss calculation. The paper presents a novel, evolutionary algorithm-based transformer optimization method which can determine the optimal conductor shape for the windings during this examined preliminary design stage. The accuracy of the presented FEM method was tested on an existing transformer design. Then the results of the proposed optimization method have been compared with a validated transformer design optimization algorithm.

This paper represents a groundbreaking paradigm shift in network optimization. Departing from traditional static methodologies, this innovative approach harnesses the power of Generative Artificial Intelligence (AI) and Large Language Models (LLMs) to optimize cloud networks dynamically. By integrating advanced AI algorithms, this framework continuously adapts and evolves, ensuring optimal real-time performance. This dynamic optimization enhances efficiency and resilience, allowing cloud networks to adjust seamlessly to changing demands and conditions. Through the fusion of cutting-edge technology and adaptive intelligence, this approach heralds a new era in network optimization, empowering organizations to achieve unprecedent-ed levels of agility and scalability in their cloud infrastructures.

A proportion of osteoarthritis (OA) patients are unsatisfied with post-operative outcomes following total joint replacement surgery (TJR), with insufficient pain relief or poor functional improvement. Predicting those will have poor outcomes would be beneficial for patients and clinicians. The aim of this study was to determine the relationship between baseline anthropometric data and the concentration of pre-operative serum and peri-operative synovial fluid (SF) cytokines with 7-month post-operative outcomes in a cohort of knee and hip OA patients. 160 OA patients were recruited who were scheduled for TJR. The concentration of 24 cytokines were measured in blood and SF by multiplex assay. EQ5Dindex health status was assessed pre-operatively and at 7-months post-operatively. 13% of patients were identified as non-responders based on EQ5Dindex. Compared to responders, non-responders were of higher BMI, had greater waist and hip circumference and had higher levels of SF leptin but lower levels of SF Resistin (p&lt;0.05). Linear regression analysis found a significant relationship between pre-operative body weight and post-operative response (ΔEQ5Dindex; r=0.222, p=0.049). The combination of body weight with SF amphiregulin and SF IL-6 provided an improved predictive model of post-operative response (r=0.470, p=0.035).

Accurate seasonal weather forecasts are vital across a broad spectrum of applications, bearing significant environmental and socioeconomic implications. This importance renders the subject a matter of primary interest to a wide range of stakeholders, including the general public, agricultural sector, emergency responders, financial institutions, and policy strategists. The need for precision in long-term predictive models necessitates the development of innovative methodologies. These methodologies should be capable of deciphering atmospheric patterns and mechanisms with detail, especially at a local level, without the resource constraints that dynamical downscaling imposes. In response to this expanding demand, this study presents a novel solution, combining a stochastic deep learning methodology, specifically Generative Adversarial Networks (GANs), with a Digital Elevation Model (DEM). The cornerstone of the proposed model is the enhancement of gridded temperature fields from seasonal forecasts. The area of focus was the Hellenic region, wherein the spatial resolution is amplified from a coarse 1° x 1° grid to an impressively detailed 0.1° x 0.1° grid. This offers a transformative perspective for interpreting and employing this crucial meteorological data. The results suggest that the downscaled fields adequately approximate the actual spatial distribution, although the predicted values tend to slightly overestimate and in some cases underestimate the original ones. This study underlines the potential of this approach to significantly enhance the resolution and utility of weather forecasts, thereby contributing to a variety of sectors dependent on reliable meteorological data.

Conditional generative adversarial networks (CGANs) learn a mapping from conditional input to observed image and perform tasks in image generation, manipulation and translation. In-situ monitoring uses sensors to obtain real-time information of additive manufacturing (AM) processes that relate to process stability and part quality. Understanding the correlations between process inputs and in-situ process signatures through machine learning can enable experimental-driven predictions of future process inputs. In this research, in-situ data obtained during a metallic powder bed fusion AM process is mapped with a CGAN. A single build of two turbine blades is monitored using EOSTATE Exposure OT, a near-infrared optical tomography system of the EOS M290 system. Layerwise images generated from the in-situ monitoring system were paired with a conditional image that labeled the specimen cross-section, laser-scan stripe overlap and z-distance to part surfaces. A CGAN was trained using the turbine blade data set and employed to generate new in-situ layerwise images for unseen conditional inputs.

Immunosenescence refers to age-related alterations of the immune system function affecting both the humoral and cellular arm of immunity. Understanding immunosenescence and its impact on vaccination of older adults is essential since primary vaccine responses in older individuals can fail to generate complete protection, especially vaccines targeting infections with increased incidence among the elderly, such as the respiratory syncytial virus. Here, we review the clinical trials of both candidate and approved vaccines against respiratory syncytial virus (RSV), which include adults aged ≥50 years, with an emphasis on the evaluation of immunogenicity parameters. Currently, there are 10 vaccine candidates and 2 vaccines approved for the prevention of RSV in the older adult population. The number of registered clinical trials for this age group amounts to 32. Our preliminary evaluation of published results and interim analyses of the RSV vaccine clinical trials indicates efficacy in older adult participants, demonstrating immunity levels that closely resemble those of younger adult participants.

The study describes the acute and delayed time course of recovery following the CrossFit® Benchmark Workout Karen. Eight trained men (28.4±6.4 years; 1RM back squat 139.1±26.0 kg) undertook the Karen protocol. The protocol consists of 150 Wall Balls, aiming to hit a target 3 meters high. Countermovement jump height (CMJ), creatine kinase (CK), and perceived recovery status scale (PRS) (general, lower and upper limbs) were assessed pre, post-0h, 24h, 48h and 72h after the session. The CK concentration 24h after was higher than pre-exercise (338.4 U/L vs. 143.3 U/L; effect size: 0.74; p≤0.05). At 48h and 72h following exercise, CK concentration had returned to baseline levels. The PRS general and of the lower limbs were lower in the 24-hours post-exercise compared to pre-exercise (PRS general: 4.7 ±1.5 and 7.9 ±1.7 mmol/L; and PRS of the lower limbs: 4.0 ±2.5 and 7.9 ±0.8, respectively). The PRS general, lower, and upper limbs were reduced at 48-post exercise compared to 72-hours post-exercise scores. Our findings provide insights into the fatigue profile and recovery in acute CrossFit® and can be useful to coaches effectively design the daily session.

Emerging semiconductor devices often lack accurate analytical models. The same is usually true of any devices working under extreme conditions like cryogenic temperatures. The usual workaround involves the use of approximation models, usually based on lookup tables or neural networks individually fitted to measurement data. In the case of experimental devices or ones working under extreme conditions, the number of units available for measurement is limited. As a result, the number of approximation-model instances is too small to enable a statistical simulation of even middle-sized circuits, which is a necessary step in integrated-circuit design since it provides the realistic picture of the circuit’s behavior in the presence of manufacturing process variations. Approximation models using structure parameters as inputs do exist in the literature, but are only useful if the end user knows the statistical distributions of those parameters, which is not usually the case. We propose a technique based on generative machine learning, namely the variational autoencoder, that uses only a small sample of devices to capture the essential features of their I–V curves under process variations and subsequently generates an arbitrary number of similarly disturbed curves. The model trained on as few as 20 instances per device type is shown to precisely reproduce the distributions of period and power consumption of a ring oscillator.

In this paper, we propose an unsupervised blind restoration model for images in hybrid degradation scenes. The proposed model encodes the content information and degradation information of images and then uses the attention module to disentangle the two kinds of information. It can improve the ability of disentangled presentation learning for a generative adversarial network (GAN) to restore the images in hybrid degradation scenes, enhance the detailed features of restored image and remove the artifact combining the adversarial loss, cycle-consistency loss, and perception loss. The experimental results on the DIV2K dataset and medical images show that the proposed method outperforms existing unsupervised image restoration algorithms in terms of peak signal-to-noise ratio (PSNR), structural similarity (SSIM), and subjective visual evaluation.

The fossil record is notorious for being incomplete and distorted, frequently conditioning the type of knowledge that can be extracted from it. In many cases, this often leads to issues when performing complex statistical analyses, such as classification tasks, predictive modelling, and variance analyses, such as those used in Geometric Morphometrics. Here different Generative Adversarial Network architectures are experimented with, testing the effects of sample size and domain dimensionality on model performance. For model evaluation, robust statistical methods were used. Each of the algorithms were observed to produce realistic data. Generative Adversarial Networks using different loss functions produced multidimensional synthetic data significantly equivalent to the original training data. Conditional Generative Adversarial Networks were not as successful. The methods proposed are likely to reduce the impact of sample size and bias on a number of statistical learning applications. While Generative Adversarial Networks are not the solution to all sample-size related issues, combined with other pre-processing steps these limitations may be overcome. This presents a valuable means of augmenting geometric morphometric datasets for greater predictive visualization.

Lignocellulosic biomass has garnered attention as an abundant and sustainable alternative energy source that can facilitate reliable and environmentally friendly energy generation. Energy security improves, and environmental impacts diminish when bioethanol, a biofuel produced via lignocellulose-based processes, is utilized. The lignocellulosic biomass of Thysanolaena latifolia has considerable potential as a bioethanol feedstock due to its high carbohydrate content (62.4 ± 0.7%). In this study, feedstock derived from T. latifolia was pretreated with various concentrations of H3PO4 to determine the ideal conditions for enzymatic hydrolysis to convert the feedstock to fermentable sugar. The findings revealed that hydrolysis efficiency and glucose recovery yields were substantially improved compared to those of the untreated sample. Pretreated samples were enzymatically digested to produce a liquid hydrolysate. This hydrolysate was fermented without detoxification using Saccharomyces cerevisiae TISTR 5339 to produce ethanol. The results indicated that T. latifolia biomass hydrolysate is a promising long-term carbon source for ethanol production from cellulosic biomass. Furthermore, the morphological and crystallographic characteristics of the treated T. latifolia biomass were influenced by H3PO4 concentration, as indicated by the SEM images, X-ray diffractogram patterns, and crystallinity index values. Therefore, utilizing T. latifolia feedstock to produce bioethanol can enhance bioenergy sustainability.

In this paper we propose a module that uses features learned by a deep convolutional neural network to infer areas of corrosion and segment pixels to corrosion areas of inspection interest. Our segmentation module is based on eigen tree decomposition and information based decision criteria. To interrogate performance we use several state-of-the-art deep learning architectures and compare to our approach. The results indicate that our method produces better results in terms of accuracy and precision, whilst maintaining good (f-score) significance over the entire dataset.

Human pose estimation is a complex detection task in which the network needs to capture the rich information contained in the images. In this paper, we propose MSTPose (Multi-Scale Transformer for human Pose estimation). Specifically, MSTPose leverages a high-resolution Convolution neural network (CNN) to extract texture information from images. For the feature maps from three different scales produced by the backbone network, each branch performs the coordinate attention operations. The feature maps are then spatially and channel-wise flattened, combined with keypoint tokens generated through random initialization, and fed into a parallel Transformer structure to learn spatial dependencies between features. As the Transformer outputs one-dimensional sequential features, the mainstream two-dimensional heatmap method is abandoned in favor of one-dimensional coordinate vector regression. The experiments show that MSTPose outperforms other CNN-based pose estimation models and demonstrates clear advantages over CNN + Transformer networks of similar types.

This study investigates the effect of perceived discrimination on the mental health of Afghan refugees, and secondly, tests the distress moderating effects of pre-migration traumatic experiences and post-resettlement adjustment factors. In a cross-sectional design, 259 Afghans completed surveys assessing perceived discrimination and a number of other factors using scales developed through inductive techniques. Multivariate analyses consisted of a series of hierarchical regressions testing the effect of perceived discrimination on distress, followed by a sequential analysis of moderator variables. Perceived discrimination was significantly associated with higher distress, and this relationship was stronger among those with a strong intra-ethnic identity, high civic engagement, and high pre-resettlement traumatic experiences. Discrimination is a significant source of stress for Afghan refugees, which may exacerbate stresses associated with other post-migration stressors. Future research is needed to tailor interventions that can help mitigate the stress associated with discrimination among this highly vulnerable group.

. Background: regular exercise elicits adaptive changes from several organs and physiological processes, including erythrocyte properties. Methods: in a group of 79 subjects (62 men and 17 women; mean age 31.37 +/- 10.19 years) who trained several times a week as they practiced amateur sports, we evaluated the elongation index, marker of erythrocyte deformability, the red blood cell distribution width (RDW), indicator of erythrocyte anisocytosis, hematocrit, hemoglobin and the main erythrocyte indices (MCV, MCH, MCHC) in basal conditions. Results: in comparison with a group of healthy not trained volunteers, the values of the elongation index, and not the RDW, are increased and this datum is accompanied by an increase of MCV and MCHC, likely related to an increased presence of circulating young erythrocytes in trained subjects. We also divided the same group according to the median of the VO2max, observing that the subgroup above the median shows both an increase in the elongation index values and a decrease in MCH and MCHC. Conclusions: in trained subjects there is no correlation between the values of elongation index and the RDW, while of particular interest and of a certain complexity seem the interrelations among elongation index, RDW and the main erythrocyte indices.

Artificial Intelligence (AI) has the power to generate scenarios and make predictions through the use of advanced algorithms and machine learning techniques. OpenAI’s GPT-3 AI is a state-of-the-art language model that has been trained on a large dataset of text, which allows it to generate hu-man-like text, and it can generate scenarios for different fields. However, GPT-3 was trained on data available up until June 2021, had no access to more recent data nor was connected to the internet. In this study, we investigated the capability of OpenAI’s GPT-3 AI to predict the Ukrainian war es-calation, which started in 2022 and had massive geopolitical effects. We used GPT-3´s capability to generate future scenarios, to check those scenarios for internal consistency, and to create a proba-bility estimate. The results showed that although GPT-3 described an open war as one of the low probability scenarios, its capability on predicting the future was limited. Furthermore, it became evident that checking internal consistency of the generated scenarios could be improved. We ap-preciated GPT-3 as very useful and powerful for generating future scenarios, but also concluded that its prediction capabilities of real-world events are limited, should be used with caution, and require further development.

Dynamic sparse training is an effective strategy to alleviate the training and inference demands of artificial neural networks. However, current sparse training methods face a challenge in achieving high levels of sparsity while maintaining performance comparable to that of their fully connected counterparts. The Cannistraci-Hebb training (CHT) method produces an ultra-sparse advantage compared to fully connected training in various tasks by using a gradient-free link regrowth method, which relies solely on the network topology. However, its rigid selection based on link prediction scores may lead to epitopological local minima, especially at the beginning of the training process when the network topology might be noisy and unreliable. In this article, we introduce the Cannistraci-Hebb training soft rule (CHTs), which applies a flexible approach to both the removal and regrowth of links during training, fostering a balance between exploring and exploiting network topology. Additionally, we investigate the network topology initialization using several approaches, including the bipartite scale-free and bipartite small-world network models. Empirical results show that CHTs can surpass the performance of fully connected networks with MLP architecture by using only 1% of the connections (99% sparsity) on the MNIST, EMNIST, and Fashion MNIST datasets and can provide remarkable results with only 0.1% of the links (99.9% sparsity). In some MLPs for image classification tasks, CHTs can reduce the active neuron network size to 20% of the original nodes (neurons), demonstrating a remarkable ability to generalize better than fully connected architectures, reducing the entire model size. This represents a relevant result for dynamic sparse training. Finally, we present evidence from larger network models such as Transformers, with 10% of the connections (90% sparsity), where CHTs outperform other prevalent dynamic sparse training methods in machine translation tasks.

The growing usage of Generative AI tools in scientific writing requires a critical examination of their benefits and challenges. This literature review is aimed at comprehensively analyzing current empirical research articles focused on the application of Generative AI in scientific writing. The Google Scholar database was used to search for the literature. The following keywords were used: "Generative AI" and "academic writing", "LLM" (Large Language Models) and "academic writing", "Generative AI" and "Scientific writing", and "ChatGPT" and "Scientific Writing". The search was restricted to articles published between January 1, 2023, and April 30, 2024. 15 articles were selected as appropriate for the study and analyzed. It was found that, thus far, ChatGPT is the most exploited tool in the studies. AI tools such as Bard (Gemini), Bing, Claude2, and Elicit were also tested. The benefits of Generative AI usage in scientific writing were found to be omnipresent. It can aid in the generation of structured abstracts, titles, introductions, literature reviews, and conclusions of a scientific article. Generative AI also makes writing more efficient and time-saving. Its capabilities in improving language and proofreading are well-established. However, the generation of inaccurate content and references by current commercially available LLMs poses a serious problem. The lack of critical thinking and tendency to produce non-original content are significant drawbacks. Generative AI should be employed with human oversight, serving as an assistant rather than a replacement. Transparency in AI usage in scientific writing is essential, along with the necessity for proper legal regulation.

This thesis presents the effects of work done on a software project for generative models and spreadsheets, allowing for a quick creation of the conceptual model of the aircraft. The subject of the work is a response to the current trends and needs prevailing in the field of computer design engineering CAD and aviation. In the initial chapters, theoretical issues related to the work being carried out were introduced and the methodology of creating software for construction and verification of the structure of aircraft along with the needs of interchange between databases of generative models was presented. In the next stages, the concepts and selected solutions for the user interface supporting the knowledge base were presented along with a set of procedures for its operation. Furthermore, the method of database integration with the methods of determining design features for the developed generative models and with the Siemens NX system. Furthermore, problems encountered in software development, as well as solution examples for model application are specified. The results obtained and the models generated on their basis were subjected to a strength analysis using Autodesk Inventor software and analysed in terms of meeting the initial assumptions. In the end, conclusions and observations resulting from the effects of the work presented in the project were formulated.

Complementary Metal-Oxide-Semiconductor (CMOS) is a typical image sensor that has a wide range of applications. However, considering the limitations of the weather condition and hardware cost, it is hard to capture high-resolution images by CMOS sensor. Recently, Super-Resolution (SR) techniques for image restoration has been gaining attentions due to its excellent performance. Under the powerful learning ability, Generative Adversarial Networks (GANs) have been proved to achieve great success. In this paper, we propose the Advanced Generative Adversarial Networks (AGAN) to efficiently correct these issues; 1) we design a Laplacian pyramid framework as pre-trained module, which is beneficial to provide multi-scale features for our input. 2) at each feature block, a convolutional skip-connections network, which may contain some latent information, is significant for generative model to reconstruct a plausible-looking image; 3) considering that edge details usually play an important role in image generation, a novel perceptual loss function is defined to train and seek optimal parameters. It is effective to achieve excellent and compelling quality captured by CMOS sensor. Quantitative and qualitative evaluations have been demonstrated that our algorithm not only fully takes advantage of Convolutional Neural Networks (CNNs) to improve the image quality, but also performs better than previous GAN algorithms for super-resolution task.

Accurate prediction of crop production is essential in effectively managing agricultural countries' food security and economic resilience. This study evaluates the performance of statistical and machine learning-based methods for large-scale crop production forecasting. We predict the quarterly production of 325 crops (including fruits, vegetables, cereals, non-food, and industrial crops) across 83 provinces in the Philippines. Using a comprehensive dataset of 10,949 time series over 13 years, we demonstrate that a global forecasting approach using a state-of-the-art deep learning architecture, the transformer, significantly outperforms traditional local forecasting approaches built on statistical and baseline methods. By leveraging cross-series information, our proposed way is scalable and works well even with time series that are short, sparse, intermittent, or exhibit structural breaks/regime shifts. The results of this study further advance the field of applied forecasting in agricultural production and provide a practical and effective decision-support tool for policymakers that oversee the farm sector on a national scale.

We propose a probabilistic graphical model for discriminative substrate characterization, to support geological and biological habitat mapping in aquatic environments. The model, called a fully connected conditional random field (CRF), is demonstrated using multispectral and monospectral acoustic backscatter from heterogeneous seafloors in Patricia Bay, British Columbia, and Bedford Basin, Nova Scotia. Unlike previously proposed discriminative machine learning algorithms, the CRF model considers both the relative backscatter magnitudes of different substrates and their relative proximities. The model therefore combines the statistical flexibility of a machine learning algorithm with an inherently spatial treatment of the substrate. The CRF model predicts substrates such that nearby locations with similar backscattering characteristics are likely to be in the same substrate class. The degree of proximity and allowable backscatter similarity are controlled by parameters that are learned from the data. CRF model results were evaluated against a popular generative model known as a Gaussian Mixture model that doesn't include spatial dependencies, only covariance between substrate backscattering response over different frequencies. Both models are used in conjunction with sparse bed observations/samples in a supervised classification. A detailed accuracy assessment, including a leave-one-out cross-validation analysis, was performed using both models. Using multispectral backscatter, the GMM model trained on 50% of the bed observations resulted in a 75% and 89% average accuracies in Patricia Bay and Bedford Basin, respectively. The same metrics for the CRF model were 78% and 95%. Further, the CRF model resulted in a 91% mean cross-validation accuracy across four substrate classes at Patricia Bay, and a 99.5% mean accuracy across three substrate classes at Bedford Basin, which suggest that the CRF model generalizes extremely well to new data. This analysis also showed that the CRF model was much less sensitive to the specific number and locations of bed observations than the generative model, owing to its ability to incorporate spatial autocorrelation in substrates. The CRF approach therefore may prove to be a powerful `spatially aware' alternative to other discriminative classifiers.

The purpose of this research was two-fold: (a) to observe whether highly trained adolescent swimmers abide to vitamin D supplement recommendations; and (b) to monitor changes in circulating 25-hydroxyvitamin D (25(OH)D) that occur between the autumn and winter months. Twenty swimmers (age: 17 ± 2 years) from a UK high-performance swimming club volunteered to complete two blood spot cards to determine 25(OH)D concentration: the first in an autumn training phase (October) and the second during winter training (January). All swimmers were advised to consume vitamin D3 supplements across the assessment period; however, only 50% of swimmers adhered to this recommendation. Resultantly, a winter decline in 25(OH)D was observed in non-supplementing swimmers (79.6 ± 25.2 to 52.6 ± 15.1 nmol·L-1, p = 0.005), with swimmers either displaying an ‘insufficient’ (60%) or ‘deficient’ (40%) vitamin D status. In comparison, greater maintenance of 25(OH)D occurred in supplementing swimmers (92.0 ± 25.5 to 97.2 ± 38.3 nmol·L-1, p = 0.544), although variable outcomes occurred at the individual level (four increased, three maintained, three declined). These findings highlight possible risks of vitamin D insufficiency during the winter for swimmers in the UK, possibly requiring standardised supplement practices. Moreover, alternative educational strategies may be required for swimmers to transfer knowledge to practice in order to improve supplement adherence in future.

The clinical requirement for a good esthetic result for immediate implant placement is the absence of dehiscence in the anterior facial alveolar bone. In the presence of a dehiscence, it is recommend-ed to use a connective tissue graft in addition to immediate implant placement or to change to ear-ly implant placement. However, the literature focusing on dehiscence is scarce, and the influence of different placement times and combined use of connective tissue graft on postoperative esthetics in cases with dehiscence is unclear. Therefore, we quantitatively evaluated the pre-extraction de-hiscence morphology and postoperative changes in the facial tissue of implants in three groups: immediate implant placement (Group I), immediate implant placement with connective tissue graft (Group IC), and early implant placement (Group E). Fifty-two implants were obtained (20 in Group I, 16 in Group IC, and 16 in Group E). A wider dehiscence increases the risk of soft tissue regression, which was one reason for choosing early implant placement. A combination of imme-diate implant placement and connective tissue graft, or early implant placement, tended to result in less soft tissue regression due to the thicker postoperative facial soft tissue volume.

The ability of deep learning to process vast data and uncover concealed malicious patterns has spurred the adoption of deep learning methods within the cybersecurity domain. Nonetheless, a notable hurdle confronting cybersecurity researchers today is the acquisition of a sufficiently large dataset to train deep learning models effectively. Privacy and security concerns associated with using real-world organization data have made cybersecurity researchers seek alternative strategies, notably focusing on generating synthetic data. Generative Adversarial Networks (GANs) have emerged as a prominent solution, lauded for their capacity to generate synthetic data spanning diverse domains. Despite their widespread use, the efficacy of GANs in generating realistic cyber attack data remains a subject requiring thorough investigation. Moreover, the proficiency of deep learning models trained on such synthetic data to accurately discern real-world attacks and anomalies poses an additional challenge that demands exploration. This paper delves into essential aspects of generative learning, scrutinizing their data generation capabilities, and conducts a comprehensive review to address the above questions. Through this exploration, we aim to shed light on the potential of synthetic data in fortifying deep learning models for robust cybersecurity applications.

Segmentation is an important step in medical imaging. In particular, machine learning, especially deep learning, has been widely used to efficiently improve and speed up the segmentation process in clinical practice. Despite the acceptable segmentation results of multi-stage models, little attention was paid to the use of deep learning algorithms for brain image segmentation, which could be due to the lack of training data. Therefore, in this paper, we propose a Generative Adversarial Network (GAN) model that performs transfer learning to segment MRI brain images.Our model enables the generation of more labeled brain images from existing labeled and unlabeled images. Our segmentation targets brain tissue images, including white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF). We evaluate the performance of our GAN model using a commonly used evaluation metric, which is Dice Coefficient (DC). Our experimental results reveal that our proposed model significantly improves segmentation results compared to the standard GAN model. We observe that our model is 2.1–10.83 minutes faster than stat-of-the-art-models.

Image segmentation is a new challenge prob- lem in medical application. The use of medical imaging has become an integral part of research, as it allows us to see inside the human body without surgical intervention. Many researcher have studied brain segmentation. One stage method is used to segment the brain tissues. In this paper, we proposed the multi-stage generative ad- versarial network to solve the problem of information loss in the one-stage. We utilize the coarse-to-fine to improve brain segmentation using multi-stage generative adversar- ial networks (GAN). In the first stage, our model generated a coarse outline for (i) background and (ii) brain tissues. Then, in the second stage, the model generated outline for (i) white matter (WM), (ii) gray matter (GM) and (iii) cerebrospinal fluid (CSF). A good result can be achieved by fusing the coarse outline and refine outline. We conclude that our model is more efficient and accu- rate in practice for both infant and adult brain segmenta- tion. Moreover, we observe that multi-stage model is faster than prior models. To be more specific, the main goal of multi-stage model is to see the performance of the model in a few shot learning case where a few labeled data are available. For medical image, this proposed model can work in a wide range of image segmentation where the convolution neural networks and one-stage methods have failed.

In this paper, our goal is to perform a virtual restoration of an ancient coin from its image. The present work is the first one to propose this problem, and it is motivated by two key promising applications. The first of these emerges from the recently recognised dependence of automatic image based coin type matching on the condition of the imaged coins; the algorithm introduced herein could be used as a pre-processing step, aimed at overcoming the aforementioned weakness. The second application concerns the utility both to professional and hobby numismatists of being able to visualise and study an ancient coin in a state closer to its original (minted) appearance. To address the conceptual problem at hand, we introduce a framework which comprises a deep learning based method using Generative Adversarial Networks, capable of learning the range of appearance variation of different semantic elements artistically depicted on coins, and a complementary algorithm used to collect, correctly label, and prepare for processing a large numbers of images (here 100,000) of ancient coins needed to facilitate the training of the aforementioned learning method. Empirical evaluation performed on a withheld subset of the data demonstrates extremely promising performance of the proposed methodology and shows that our algorithm correctly learns the spectra of appearance variation across different semantic elements, and despite the enormous variability present reconstructs the missing (damaged) detail while matching the surrounding semantic content and artistic style.

Cooling fans are one of the critical components of air-forced (AF) dry-type transformers for regulating internal temperatures. Therefore, effective malfunction detection is crucial to maintain the transformer temperature within an acceptable range and prevent overheating. Regular maintenance occurs periodically and issues with cooling fans may arise between maintenance periods, leading to prolonged operation under malfunctioning conditions and potential failures. In addition, utilities typically have online information about whether a fan works or not without providing information about cooling fan malfunctioning circumstances. To address these challenges, this study proposes learning-based online monitoring techniques to detect malfunctions in AF transformer cooling fans. Random forests (RFs) and convolutional neural networks (CNNs) are developed to classify the audio signals from cooling fans into normal and malfunctioning classes. Unlike RFs, which require separate feature extraction, CNNs are trained based on spectrogram images derived from audio signals. Thus, various time-frequency techniques are utilized for feature extraction in RFs. Besides, multiple data augmentation techniques are employed to enhance the dataset size and diversity. Algorithmic performance is optimized through hyperparameter tuning and classifier threshold adjustment. Simulations reveal that CNNs outperform RFs, whereas the latter provides superior interpretability of acoustic features compared to the former.

Deep learning technologies for skin cancer detection have been dramatically advanced based on high resolution dermoscopic images. The low-cost approach based on clinical skin images of low resolution is promising but remains technically challenging due to their undermined image quality. In this paper, we propose a coarse-to-fine efficient super resolution transformer (CF-ESRT) network to reconstruct the dermoscopy-level high resolution skin image from a low resolution clinical image. By connecting the refinement network to the original super resolution transformers and applying perceptual and gradient losses, our framework noticeably improves the finer texture details of skin lesions in the super resolution (SR) images, and is effective to elevate the perceptual quality of the SR images. Quantitative and qualitative evaluations show that our method outperforms ESRT the basis model as well as the other state-of-the-art SR models.

In order to achieve the target, several activities are being undertaken, including the development of multipurpose reservoir planning for the water supply and sanitation infrastructure, irrigation, and hydropower development for the H project. To undertake this work, the main data given was rainfall data, flow data, catchment area, base population for water supply, and so on. This report presents a specifically preliminary study and design of water supply, irrigation, and hydropower development for the project, which is carried out in accordance with the scope of work given by our professor. The preliminary study and design are therefore to investigate scenarios and alternatives and prepare a document of water supply, irrigation, and hydropower infrastructures for the provision of the assignment. This report, as required in the assignment, contains project background, water sources assessment, selection of suitable hydropower potential, and irrigation sizing, design criteria, descriptions of the proposed scheme, preliminary design of water supply components. The objective of the project is to recommend and design a cost-effective water supply system, irrigation, and hydropower development to supply reliable, sufficient, and safe water from a defined water source to the H project. The specific objectives of the investigation works are: • To assess potential water source alternatives for water supply • Selecting appropriate size for the reservoir which could satisfy all water demand • Assess and select suitable irrigation development and hydropower potential. • Preliminary study and design of all components of the reservoir and appurtenant structures The scope of the project generally focuses on preliminary study and design of water supply, irrigation and hydropower development components from water sources potential and the water balance output.

Proposed in 1889 by Svante Arrhenius, the Arrhenius equation is an important result in physical chemistry which aims to find the relationship between temperature and reaction rates. It is also an essential equation for analytic chemistry which helps to determine the energy barrier for a given reaction. In this article, we will go through a detailed derivation of the exponential factor in the Arrhenius Equation based on the Boltzmann distribution of particle energy probability. Then, we will attempt to model the pre-exponential factor in accordance with the principles of collision theory for bi-molecular reaction with gaseous reactants. Understanding the basic principles of chemical reaction provides valuable tools to analyze complex reaction mechanisms in the perspective of optimizing existing industrial processes.

Body Expression (BE) has been defined in the past decades as a discipline within Physical Education (PE) with very particular characteristics and a strong emotional component. In this study, a programme of this discipline was applied to university Physical Activity and Sports Sciences (PASS) students from six consecutive academic years: three prior to and three during the pandemic. A pre-post design was used to determine how the BE programme affected the university students' self-concept (SC). To do so, a questionnaire with a multidimensional approach to this construct was administered, whose dimensions were closely related to the BE programme's characteristics. The results revealed significant improvements in the final SC, compared to the initial SC. Men reported lower SC values than women before the programme implementation, but higher at the end. Therefore, the change was larger in men, so the programme may have had an equalising effect between groups. It was also verified that the pandemic had particularly affected women.

The accurate prediction of the solar Diffuse Fraction (DF), sometimes called the Diffuse Ratio, is an important topic for solar energy research. In the present study, the current state of Diffuse Irradiance research is discussed and then three robust, Machine Learning (ML) models, are examined using a large dataset (almost 8 years) of hourly readings from Almeria, Spain. The ML models used herein, are a hybrid Adaptive Network-based Fuzzy Inference System (ANFIS), a single Multi-Layer Perceptron (MLP) and a hybrid Multi-Layer Perceptron-Grey Wolf Optimizer (MLP-GWO). These models were evaluated for their predictive precision, using various Solar and Diffuse Fraction (DF) irradiance data, from Spain. The results were then evaluated using two frequently used evaluation criteria, the Mean Absolute Error (MAE) and the Root Mean Square Error (RMSE). The results showed that the MLP-GWO model, followed by the ANFIS model, provided a higher performance, in both the training and the testing procedures.

This paper applies a theoretical model to analyze the ways that widespread use of generative AIs (GAIs) in education and, more broadly, in contributing to and reflecting the collective intelligence of our species, can and will change us. The model extends Brian Arthur’s insights into the nature of technologies as the orchestration of phenomena to our use by explaining the nature of humans participation in their enactment, whether as part of the orchestration (hard technique, where our roles must be performed correctly) or as orchestrators of phenomena (soft technique performed creatively or idiosyncratically). Education may be seen as a technological process for developing the soft and hard techniques of humans to participate in the technologies and thus the collective intelligence of our cultures. Unlike all earlier technologies, by embodying that collective intelli-gence themselves, GAIs can closely emulate and implement not only the hard technique but also the soft that, until now, was humanity’s sole domain: the very things that technologies enabled us to do can now be done by the technologies themselves. The consequences for what, how, and even whether we learn are profound. The paper explores some of these consequences and concludes with theoretically informed approaches that may help us to avert some dangers while benefiting from the strengths of generative AIs.

The use of visual signals in horticulture has attracted significant attention and encompassed a wide range of data types such as 2D images, videos, hyperspectral images, and 3D point clouds. These visual signals have proven to be valuable in developing cutting-edge computer vision systems for various applications in horticulture, enabling plant growth monitoring, pest and disease detection, quality and yield estimation, and automated harvesting. However, unlike other sectors, developing deep learning computer vision systems for horticulture encounters unique challenges due to the limited availability of high-quality training and evaluation datasets necessary for deep learning models. This paper investigates the current status of vision systems and available data in order to identify the high-quality data requirements specific to horticultural applications. We analyse the impact of the quality of visual signals on the information content and features that can be extracted from these signals. To address the identified data quality requirements, we explore the usage of a deep learning-based super-resolution model for generative quality enhancement of visual signals. Furthermore, we discuss how these can be applied to meet the growing requirements around data quality for learning-based vision systems. We also present a detailed analysis of the competitive quality generated by the proposed solution compared to cost-intensive hardware-based alternatives. This work aims to guide the development of efficient computer vision models in horticulture by overcoming existing data challenges and paving a pathway forward for contemporary data acquisition.

When integrating data from different sources, there are problems of synonymy, different languages, concepts of different granularity. This paper proposes a simple but effective approach to evaluate the semantic similarity of short texts, especially keywords. The method is capable of matching keywords from different sources and languages by exploiting transformers and WordNet-based methods. Key features of the approach include its unsupervised pipeline, mitigation of the lack of context in keywords, scalability for large archives, support for multiple languages and real-world scenarios adaptation capabilities. The work aims to provide a versatile tool for different cultural heritage archives without requiring complex customization. The objectives of the paper are to explore different approaches to identifying similarities in 1- or n-gram tags, to evaluate and compare different pre-trained language models, and to define integrated methods to overcome limitations. Tests to validate the approach have been conducted using the QueryLab portal, a search engine for cultural heritage archives, to evaluate the proposed pipeline.

In the modern era of genomic research, the scientific community is witnessing an explosive growth in the volume of published findings. While this abundance of data offers invaluable insights, it also places a pressing responsibility on genetic professionals and researchers to stay informed about the latest findings and their clinical significance. Genomic variant interpretation is currently facing a challenge in identifying the most up-to-date and relevant scientific papers, while also extracting meaningful information to accelerate the process from clinical assessment to reporting. Computer-aided literature search and summarization can play a pivotal role in this context. By synthesizing complex genomic findings into concise, interpretable summaries, this approach facilitates the translation of extensive genomic datasets into clinically relevant insights. To bridge this gap, we present VarChat (varchat.engenome.com), an innovative tool based on generative AI, developed to find and summarize the fragmented scientific literature associated with genomic variants into brief yet informative texts.VarChat provides users with a concise description of specific genetic variants, detailing their impact on related proteins and possible effects on human health. Additionally, VarChat offers direct links to related scientific trustable sources, and encourages deeper research. Availability: VarChat is freely available at varchat.engenome.com

Image fusion is the process of combining multiple input images from single or multiple imaging modalities into a fused image, which is expected to be more informative for human or machine perception as compared to any of the input images. In this paper, we propose a novel method based on deep learning for fusing infrared images and visible images, named the LBP-based proportional input generative adversarial network (LPGAN). In the image fusion task, the preservation of structural similarity and image gradient information is contradictory, and it is difficult for both to achieve good performance at the same time. To solve this problem, we innovatively introduce Local Binary Patterns (LBP) into Generative Adversarial Networks (GANs), which effectively utilize the texture features of the source images, so that the network has stronger feature extraction ability and anti-interference ability. In the feature extraction stage, we introduce a pseudo-siamese network for the generator to extract the detailed features and the contrast features. At the same time, considering the characteristic distribution of different modal images, we propose a 1:4 scale input mode. Extensive experiments on the publicly available TNO dataset and CVC14 dataset show that the proposed method achieves the state-of-the-art performance. We also test the universality of LPGAN through the fusion of RGB and infrared images on the RoadScene dataset. In addition, LPGAN is applied to multi-spectral remote sensing image fusion. Both qualitative and quantitative experiments demonstrate that our LPGAN can not only achieve good structural similarity, but also retain rich detailed information.

Nowadays, we use social networks such as Twitter, Facebook, WeChat and Weibo as means to communicate with each other and get to know others. Gradually social networks has become indispensable in our everyday life, and we cannot absolutely imagine what the daily life would be like without social networks. Through social networks, we can access friends' opinions and behaviors easily and are influenced by them in turn. Thus, an effective algorithm to find the top-K influential nodes (the problem of influence maximization) in the social network is critical for various downstream tasks such as viral marketing, anticipating natural hazards, reducing gang violence, public opinion supervision etc. Solving the problem of influence maximization in real-world propagation scenarios often involves estimating influence strength (influence probability between two nodes), which cannot directly observed. To estimate influence strength, conventional approaches propose various humanly-devised rules to extract features of user interactions, the effectiveness of which heavily depends on domain expert knowledge. Besides, they are often applicable for special scenarios or specific diffusion models. Consequently, they are difficult to be generalized into different scenarios, diffusion models and even domains. Inspired by the powerful ability of neural networks in the field of representation learning, we design a deep hierarchical network embedding model HGE to map nodes (with attributes) into latent space automatically. In general, HGE takes an attributed social network as the input for learning latent network representation of each node, incorporating hierarchical community structure, node attributes and general network structure into a unified deep generative framework. Then, with the leaned latent representation of each node, we propose a HGE-GA algorithm to predict influence strength and compute the top-K influential nodes through a greedy-based maximization algorithm. Extensive experiments on real-world attributed networks demonstrate the outstanding superiority of the proposed HGE model and HGE-GA algorithm compared with the state-of-the-art methods, verifying the effectiveness of the proposed model and algorithm.

Nowadays, and with the mechanization of life, speech processing has become so crucial for the interaction between humans and machines. Deep neural networks require a database with enough data for training. The more features are extracted from the speech signal, the more samples are needed to train these networks. Adequate training of these networks can be ensured when there is access to sufficient and varied data in each class. If there is not enough data; it is possible to use data augmentation methods to obtain a database with enough samples. One of the obstacles to developing speech emotion recognition systems is the Data sparsity problem in each class for neural network training. The current study has focused on making a cycle generative adversarial network for data augmentation in a system for speech emotion recognition. For each of the five emotions employed, an adversarial generating network is designed to generate data that is very similar to the main data in that class, as well as differentiate the emotions of the other classes. These networks are taught in an adversarial way to produce feature vectors like each class in the space of the main feature, and then they add to the training sets existing in the database to train the classifier network. Instead of using the common cross-entropy error to train generative adversarial networks and to remove the vanishing gradient problem, Wasserstein Divergence has been used to produce high-quality artificial samples. The suggested network has been tested to be applied for speech emotion recognition using EMODB as training, testing, and evaluating sets, and the quality of artificial data evaluated using two Support Vector Machine (SVM) and Deep Neural Network (DNN) classifiers. Moreover, it has been revealed that extracting and reproducing high-level features from acoustic features, speech emotion recognition with separating five primary emotions has been done with acceptable accuracy.