wjaskowski/TCIAIG-2015-Co-CMA-ES
Coevolutionary CMA-ES for Knowledge-Free Learning of Game Position Evaluation
Performance profiles generation for Weighted Piece Counter (WPC) strategies of Othello
Public repository record indexed from GitHub. Explore verified star velocity metrics, source code repositories, and curated developer tool directories across the GitHubRepo ecosystem.
Coevolutionary CMA-ES for Knowledge-Free Learning of Game Position Evaluation
Code repository for GECCO 2015 paper: "N-Tuple Network for Knowledge-Free Reinforcement Learning in High Dimensions: A Case Study in SZ-Tetris"
Game 2048 GECCO 2015 Competition
Code for Artificial and Computational Intelligence Methods Course at Poznan University of Technology, Poland