Random Forests for Opponent Hand Estimation in Gin Rummy

Authors

  • Anthony Hein Princeton University
  • May Jiang Princeton University
  • Vydhourie Thiyageswaran Princeton University
  • Michael Guerzhoy Princeton University University of Toronto Li Ka Shing Knowledge Institute

DOI:

https://doi.org/10.1609/aaai.v35i17.17830

Keywords:

Gin Rummy, Guerzhoy@princeton.edu, Random Forsts, Supervised Learning

Abstract

We demonstrate an AI agent for the card game of Gin Rummy. The agent uses simple heuristics in conjunction with a model that predicts the probability of each card's being in the opponent's hand. To estimate the probabilities for cards' being in the opponent's hand, we generate a dataset of Gin Rummy games using self-play, and train a random forest on the game information states. We explore the random forest classifier we trained and study the correspondence between its outputs and intuitively correct outputs. Our agent wins 61% of games against a baseline heuristic agent that does not use opponent hand estimation.

Downloads

Published

2021-05-18

How to Cite

Hein, A., Jiang, M., Thiyageswaran, V., & Guerzhoy, M. (2021). Random Forests for Opponent Hand Estimation in Gin Rummy. Proceedings of the AAAI Conference on Artificial Intelligence, 35(17), 15545-15550. https://doi.org/10.1609/aaai.v35i17.17830