policy

Recursive-Learning-Chess

alcsel · JAX

or hover any field below to flag it

Overview

Name
Recursive-Learning-Chess
Author
alcsel
Framework
JAX
License
unknown
Skill type
other
Evidence level
untested
Task description
Chess RL agent trained via self-play using JAX/Flax on PGX environment. Features 8 residual blocks, actor-critic architecture, and epsilon-greedy exploration with a frozen opponent updated every 500 batches. Trained on Kaggle T4 with 2048 parallel environments.

Spaces

Action space
other · 0-dim · 0Hz
Observation space
  • type: other

Links

HuggingFace repo
null
Paper (arXiv)
null

Compatible robots

3+17 mentioned but not in catalog yet

Compatible environments

0

No environments list Recursive-Learning-Chess yet.

Datasets that reference this policy

0

No datasets reference Recursive-Learning-Chess yet.