Predicting Chess Puzzle Difficulty with Transformers

TL;DR

This paper introduces GlickFormer, a transformer-based model that predicts chess puzzle difficulty by modeling human perception and complexity, outperforming existing methods and aiding personalized chess training.

Contribution

The paper presents GlickFormer, a novel transformer architecture that combines spatial and temporal features to accurately predict chess puzzle difficulty, bridging game theory and human cognition.

Findings

GlickFormer outperforms the ChessFormer baseline in difficulty prediction metrics.

The model achieved 11th place in the IEEE BigData 2024 Cup competition.

Experimental results validate the effectiveness of spatio-temporal modeling for difficulty assessment.

Abstract

This study addresses the challenge of quantifying chess puzzle difficulty - a complex task that combines elements of game theory and human cognition and underscores its critical role in effective chess training. We present GlickFormer, a novel transformer-based architecture that predicts chess puzzle difficulty by approximating the Glicko-2 rating system. Unlike conventional chess engines that optimize for game outcomes, GlickFormer models human perception of tactical patterns and problem-solving complexity. The proposed model utilizes a modified ChessFormer backbone for spatial feature extraction and incorporates temporal information via factorized transformer techniques. This approach enables the capture of both spatial chess piece arrangements and move sequences, effectively modeling spatio-temporal relationships relevant to difficulty assessment. Experimental evaluation was conducted on a dataset of over 4 million chess puzzles. Results demonstrate GlickFormer's superior performance compared to the state-of-the-art ChessFormer baseline across multiple metrics. The algorithm's performance has also been recognized through its competitive results in the IEEE BigData 2024 Cup: Predicting Chess Puzzle Difficulty competition, where it placed 11th. The insights gained from this study have implications for personalized chess training and broader applications in educational technology and cognitive modeling.