Learning Chess With Language Models and Transformers

TL;DR

This paper explores using language models and transformers to understand and learn board games like Nim and chess, demonstrating that models can grasp game rules and compete at high levels.

Contribution

It introduces a novel approach of applying BERT-based models to game learning, including Nim and chess, showing their ability to learn rules and perform competitively.

Findings

Models learned Nim game rules despite noise

Language models achieved high-level chess performance

Models can compete against Stockfish at category-A rating

Abstract

Representing a board game and its positions by text-based notation enables the possibility of NLP applications. Language models, can help gain insight into a variety of interesting problems such as unsupervised learning rules of a game, detecting player behavior patterns, player attribution, and ultimately learning the game to beat state of the art. In this study, we applied BERT models, first to the simple Nim game to analyze its performance in the presence of noise in a setup of a few-shot learning architecture. We analyzed the model performance via three virtual players, namely Nim Guru, Random player, and Q-learner. In the second part, we applied the game learning language model to the chess game, and a large set of grandmaster games with exhaustive encyclopedia openings. Finally, we have shown that model practically learns the rules of the chess game and can survive games against Stockfish at a category-A rating level.