Chess AI: Competing Paradigms for Machine Intelligence

Abstract

Endgame studies have long served as a tool for testing human creativity and intelligence. We find that they can serve as a tool for testing machine ability as well. Two of the leading chess engines, Stockfish and Leela Chess Zero (LCZero), employ significantly different methods during play. We use Plaskett's Puzzle, a famous endgame study from the late 1970s, to compare the two engines. Our experiments show that Stockfish outperforms LCZero on the puzzle. We examine the algorithmic differences between the engines and use our observations as a basis for carefully interpreting the test results. Drawing inspiration from how humans solve chess problems, we ask whether machines can possess a form of imagination. On the theoretical side, we describe how Bellman's equation may be applied to optimize the probability of winning. To conclude, we discuss the implications of our work on artificial intelligence (AI) and artificial general intelligence (AGI), suggesting possible avenues for future research.

arXiv comment15 pages, 8 figures

References in corpus (8)

• Neural Machine Translation by Jointly Learning to Align and Translate
• Distilling the Knowledge in a Neural Network
• Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model
• Learning both Weights and Connections for Efficient Neural Networks
• A Game-Based Approximate Verification of Deep Neural Networks with Provable Guarantees
• Global optimization of quantum dynamics with AlphaZero deep exploration
• Towards "AlphaChem": Chemical Synthesis Planning with Tree Search and Deep Neural Network Policies
• The Digital Synaptic Neural Substrate: A New Approach to Computational Creativity

Cited by in corpus (4)

• Learning Chess With Language Models and Transformers
• The Value of Chess Squares
• Impartial Games: A Challenge for Reinforcement Learning
• Gambits: Theory and Evidence