conftrace_

2021 AAAI AAAI 2021

A Data-Driven Approach for Gin Rummy Hand Evaluation

Abstract

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.

🌉 Interdisciplinary Bridge - Artificial Intelligence and Reinforcement Learning

Authors

Sang T. Truong, Todd W. Neller

Topics

Machine Learning > Core Methods > Classification Machine Learning > Optimization & Theory > Bayesian Inference Reinforcement Learning > Applications > Game AI Machine Learning > Bayesian & Probabilistic > Probabilistic Modeling Deep Learning > Architectures > Convolutional Neural Networks

Keywords

monte carlo simulation convolutional neural network bayesian reasoning game state hand evaluation offensive scoring

More from AAAI 2021

All AAAI 2021 papers →

A Deterministic Neural Network Approach to Playing Gin Rummy

Introduction to Machine Learning with Robots and Playful Learning

Designing a Hybrid AI Residency

Teaching Tech to Talk: K-12 Conversational Artificial Intelligence Literacy Curriculum and Development Tools

GANs Unplugged

Educational Question Mining At Scale: Prediction, Analysis and Personalization