Connect X - RL Kaggle

Project information

Connect Four agent combining a Policy–Value CNN with Negamax search under a 2-second move budget. Experiments cover reinforcement learning, supervised learning from expert games, self-play, and search caching, with versioned data and configurable training pipelines.

Python PyTorch kaggle connect4 reinforcemen... supervised l... machine lear... adversarial-... negamax

77 Commits
1 Branch
0 Tags

README
MIT License

Created on

September 28, 2026

Loading