Guilherme graduated with a Bachelor's degree in Science and Technology in 2019, a Bachelor's degree in Computer Science in 2020, and a Bachelor's degree in Neuroscience in 2023, all graduations from the Federal University of ABC (UFABC), Santo Andre, Brazil.
He entered the master's program in early 2020, also at UFABC, and obtained his master's degree in Computer Science in early 2022. He has been pursuing a doctoral degree since 2022, also at the same institution. In 2023, he was awarded a Doctorate Sandwich grant from Coordination of Superior Level Staff Improvement (CAPES)—Brazil and went to Boston Children's Hospital/Harvard Medical School to join the cavalab for six months.
Since 2017, he has studied evolutionary algorithms and symbolic regression, his area of greatest expertise. In 2020, he studied interpretability in machine learning, focusing on applying explanation methods in the context of symbolic regression. During his undergraduate studies, he also studied digital signal processing and sound source localization, the processing of functional magnetic resonance imaging images, and functional connectivity analysis through graph theory.
He is currently interested in evolutionary computing, interpretability in machine learning, optimization methods for symbolic regression, and computational models of neural information processing.
You can find a list of all my publications with available source code in my GitHub below, ordered chronologically (from first to last).
My research
- Lightweight symbolic regression with the interaction-transformation representation (📂 github repo)
- A Parametric Study of Interaction-Transformation Evolutionary Algorithm for Symbolic Regression (📂 github repo)
- Interaction–Transformation Evolutionary Algorithm for Symbolic Regression (📂 github repo)
- Measuring feature importance of symbolic regression models using partial effects (📂 github repo)
- Interpretability in symbolic regression: a benchmark of explanatory methods using the Feynman data set (📂 github repo)
- Interaction-transformation evolutionary algorithm with coefficients optimization (📂 github repo)
- Inexact Simplification of Symbolic Regression Expressions with Locality-sensitive Hashing (📂 github repo 📃 arXiv)
- Minimum variance threshold for 𝜖-lexicase selection (📂 github implementation 📂 github srbench experiments 📃 arXiv)
I've learned several languages throughout my academic life, some better than others. My main languages now are Julia and Python, the ones that I have a greater domain. Throughout my academic life, I also used other languages in my projects, such as R, C++, javascript, and Java.
As for my GitHub repositories, there is some statistics about the languages I've used:
Last Edited on: 05/12/2022