gaetanserre Goto Github PK

followers: 23.0 following: 15.0 repos: 36.0 gists: 0.0

Name: Gaëtan Serré

Type: User

Company: Centre Borelli, ENS Paris-Saclay

Bio: PhD Student in applied mathematics

Location: Gif-Sur-Yvette

Blog: gaetanserre.fr

Hello there!

My name is Gaëtan Serré, I am a PhD student in mathematics at Centre Borelli, the math lab of the École Normale Supérieure de Paris-Saclay.

I work on

measure/probability theory;
stochastic global optimization;
flows.

Note: Top languages is only a metric of the languages my public code consists of and doesn't reflect experience or skill level.

Main projects

Global Optimization: Implementation of several global optimization algorithms. See demo here.
LIPO-E: Set of global optimization algorithms derivated from Malherber and Vayatis, 2017. See paper.
SVGD analysis: An implementation of the Stein Variational Gradient Descent algorithm along with a detailed summary of the theoretical results.
GAiA: A chess program that uses a complex residual neural network specialized in image recognition.
LiSA: A 3D path tracing render engine.
L2RPN 2022 PPO Baseline: The code of the baseline agent provided in the 2022 edition of Learning to Run a Power Network.
AlphaZero (RLlib): An implementation of the AlphaZero algorithm in a fork of the RLlib library.

CV

CV (EN)

Connect with me

Gaëtan Serré's Projects

aes-electron

File encryption and PDF digital signature software made with Electron.

aes_256-python

An AES-256 file encryption tool written in Python 3

aoc

Repository for my solutions to the Advent of Code 2023 challenges in Lean 4.

chess-viewer

A easy-to-use GUI for chess engine powered by electron.

chessanalyzer

An in-depth chess game analyzer written with Python 3 using Stockfish engine

formal-language-algorithms

Algorithms for formal language written with Python 3 with nltk

gaia

GAiA is a UCI chess engine built with C++ 17, ONNX and Pytorch. It performs an in-depth analysis and uses a complex squeeze-and-excitation residual network to evaluate each chess board.

kawa

Compilateur du langage objet Kawa vers l'assembleur MIPS en OCaml.

This repository contains the code to train the baseline agent provided in the 2022 edition of Learning to Run a Power Network and to recreate the experiments (as well as the figures) of the paper Reinforcement learning for Energies of the future and carbon neutrality: a Challenge Design.