A Data-Driven Approach for Gin Rummy Hand Evaluation

Authors

  • Sang T. Truong DePauw University
  • Todd W. Neller Gettysburg College

Keywords:

Gin Rummy, Convolutional Neural Networks, Pattern Recognition, Bayes' Rule, Probabilistic Reasoning, Monte Carlo Simulation, Imperfect Information Games

Abstract

We develop a data-driven approach for hand strength evaluation in the game of Gin Rummy. Employing Convolutional Neural Networks, Monte Carlo simulation, and Bayesian reasoning, we compute both offensive and defensive scores of a game state. After only one training cycle, the model was able to make sophisticated and human-like decisions with a 55.4% +/- 0.8% win rate (90% confidence level) against a Simple player.

Downloads

Published

2021-05-18

How to Cite

Truong, S. T., & Neller, T. W. (2021). A Data-Driven Approach for Gin Rummy Hand Evaluation. Proceedings of the AAAI Conference on Artificial Intelligence, 35(17), 15647-15654. Retrieved from https://ojs.aaai.org/index.php/AAAI/article/view/17843