The Influence of the Size of the Search Space on Learning to Play Chess using Deep Reinforcement Learning Algorithms

The Influence of the Size of the Search Space on Learning to Play Chess using Deep Reinforcement Learning Algorithms

Hakim Zakuto, Aksel (TU Delft Electrical Engineering, Mathematics and Computer Science)

de Weerdt, M.M. (mentor)
Neustroev, G. (mentor)

Delft University of Technology

Computer Science and Engineering

Benchmarking Deep Reinforcement Learning

2021-06-28

The current state-of-the-art solutions for playing Chess, are created using deep reinforcement learning. AlphaZero, the current world champion, uses ’policy networks’ and ’value network’ for selecting moves and evaluating positions respectively. However, the training of these networks are done using reinforcement learning from games of selfplay. There are many factors which determine the learning speed of reinforcement learning algorithms, where the size of the search space is a main one. In this research, we have tried to see the effect of the size of the search space on the time it takes the reinforcement learning agent to learn.

Rinforcement Learning
Chess
Search Space

http://resolver.tudelft.nl/uuid:a035f5f9-e875-42ed-b448-fcdeeda6c3d8

Student theses

bachelor thesis

© 2021 Aksel Hakim Zakuto