Mastering Gomoku with AlphaZero: A Study in Advanced AI Game Strategy

Authors

  • Wen Liang University of California, San Diego
  • Chao Yu University of California, San Diego
  • Brian Whiteaker Brian Whiteaker University of California, San Diego La Jolla, CA 92093 bwhiteak@ucsd.edu
  • Inyoung Huh University of California, San Diego La Jolla, CA 92093 i1huh@ucsd.edu
  • Hua Shao University of California, San Diego
  • Youzhi Liang

Abstract

This study delves into the application of the AlphaZero algorithm to Gomoku, a classic board game. Unlike traditional AI methods, AlphaZero learns and strategizes without human input. Our research contrasts AlphaZero's innovative approach with the Monte Carlo tree search technique, highlighting its advanced capabilities in strategic decision-making. The findings reveal AlphaZero's remarkable proficiency in mastering the complexities of Gomoku, marking a significant advancement in artificial intelligence's role in game strategy and decision-making. This paper provides a comprehensive analysis of AlphaZero's learning process and strategic execution in Gomoku, offering insights into the future of AI in strategic gaming.

Mastering Gomoku with AlphaZero

Downloads

Published

2023-11-12

How to Cite

Liang, W., Yu, C., Whiteaker, B., Huh, I., Shao, H., & Liang, Y. (2023). Mastering Gomoku with AlphaZero: A Study in Advanced AI Game Strategy. Sage Science Review of Applied Machine Learning, 6(11), 32–43. Retrieved from https://journals.sagescience.org/index.php/ssraml/article/view/115