Evolving Agents for the Hanabi 2018 CIG Competition

TL;DR

This paper presents a genetic algorithm to evolve rule-based agents for Hanabi, improving performance in a competitive setting by optimizing rule sequences and expanding rule sets across different game sizes.

Contribution

It introduces a genetic algorithm approach for evolving Hanabi agents, including rule set expansion and specialization for various game sizes, outperforming previous methods.

Findings

Achieved superior scores in Hanabi agent competitions.

Demonstrated effectiveness of evolved rule sequences.

Showed benefits of rule set expansion and specialization.

Abstract

Hanabi is a cooperative card game with hidden information that has won important awards in the industry and received some recent academic attention. A two-track competition of agents for the game will take place in the 2018 CIG conference. In this paper, we develop a genetic algorithm that builds rule-based agents by determining the best sequence of rules from a fixed rule set to use as strategy. In three separate experiments, we remove human assumptions regarding the ordering of rules, add new, more expressive rules to the rule set and independently evolve agents specialized at specific game sizes. As result, we achieve scores superior to previously published research for the mirror and mixed evaluation of agents.