Bot Pokémon Gen 1 (Random Battle Showdown) entraîné en RL avec MaskablePPO, imitation + curriculum + self-play ; suivi MLflow.
Pokemon Reinforcement Learning with RLlib
Local Pokémon Showdown + LLM battle agents: web manager for matches and tournaments, replays and stats, optional broadcast/Twitch streaming.
A DSL for Pokémon battle policies (Rust compiler, Python runtime, MCP server), built to measure which steering strategy best teaches a coding agent a language it has never seen. Graded by real battle wins, not an LLM judge.
Pokémon battle agent using LLM with poke-env, planned to extend with reinforcement learning.
Competitive Pokémon Showdown ML battle agent: BC → offline RL → PPO self-play across gen9randombattle, gen9ou and gen9vgc2025regi, with every claim gated on a pre-registered evaluation.
Autonomous, game-theoretic Pokémon Showdown AI powered by Simultaneous Expectiminimax, Set-Transformer neural policy pruning (ONNX), inverse damage calculations, and Smogon metagame priors.
To associate your repository with the poke-env topic, visit your repo's landing page and select "manage topics."