Chess AI Training Data with Zone Database and MCVS Engine

A repository containing a reinforcement learning agent and training infrastructure for Chess, implementing a Monte-Carlo Value Search (MCVS) engine with a Displacement-based ABC Model and Weighted Adjacency Matrices. The dataset includes a trained Zone Database for position evaluation, uploaded by author chaim-duchovny to Hugging Face and last updated on 2026-04-12.

Use Cases

• Training reinforcement learning agents for Chess based on the described Monte-Carlo Value Search engine.
• Evaluating chess positions using the provided trained Zone Database.
• Benchmarking novel AI models against the described Displacement-based ABC Model and Weighted Adjacency Matrices.
• Studying the application of homogeneous coordinates in game AI for board state representation.

Strengths

• Implements a novel Displacement-based ABC Model with homogeneous coordinates.
• Includes a trained Zone Database specifically for chess position evaluation.
• Last updated on 2026-04-12, indicating recent maintenance.

Limitations

• Description metadata is limited; actual data quality, column definitions, and row count require manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred from the model description or code.

Provenance

Source

huggingface

Collection Method

Likely generated through custom reinforcement learning training infrastructure.

Time Range

null

Freshness

Last updated 2026-04-12 21:35:34.

Geography

null