Paper ID: 2409.12272

Mastering Chess with a Transformer Model

Daniel Monroe, The Leela Chess Zero Team

Transformer models have demonstrated impressive capabilities when trained at scale, excelling at difficult cognitive tasks requiring complex reasoning and rational decision-making. In this paper, we explore the application of transformer models to chess, focusing on the critical role of the position encoding within the attention mechanism. We show that in chess, transformers endowed with a sufficiently versatile position encoding can match existing chess-playing models at a fraction of the computational cost. Our architecture significantly outperforms AlphaZero at 8x fewer FLOPS and matches prior grandmaster-level transformer-based agents at 30x fewer FLOPS.

Submitted: Sep 18, 2024