You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This repository showcases a project where the NEAT algorithm, which dynamically evolves neural networks by adjusting both their architecture and weights, is applied to develop an advanced game-playing agent. The project demonstrates NEAT's capability to create optimized strategies for complex gameplay scenarios.
This repository develops an AlphaZero-inspired agent for Quixo, leveraging deep learning and Monte Carlo Tree Search (MCTS) for strategy enhancement. It emphasizes self-play, dynamic data management, and exploration-exploitation balance in MCTS, aiming for peak performance in Quixo.
Implémentation en C du jeu de plateau Quixo. Quixo est un jeu de pions abstrait dérivé du "Tic-Tac-Toc" (jeu d'alignement). Le matériel se compose d'un plateau et de 25 cubes.