Accession Number:

AD1166929

Title:

Exploring Learning Classifier System Behaviors in Multi-Action, Turn-based Wargames

Descriptive Note:

[Technical Report, Master's Thesis]

Corporate Author:

AIR FORCE INSTITUTE OF TECHNOLOGY WRIGHT-PATTERSON AFB OH

Personal Author(s):

• Terlizzi, Garth J.

Report Date:

2022-03-24

Pagination or Media Count:

102

Abstract:

State of the art game-playing Artificial Intelligence research focuses heavily on non-symbolic learning methods. These methods offer little explainable insight into their decision-making processes. Learning Classifier Systems LCSs provide an alternative. LCSs use rule-based learning, guided by a Genetic Algorithm GA, to produce a human-readable rule-set. This thesis explores LCS usefulness in game-playing agents for multi-agent wargames. Several Multi-Agent Learning Classifier System MALCS variants are implemented in the wargame Stratagem MIST a Zeroeth-Level Classifier System ZCS, an extended Classifier System XCS, and an Adaptive Pittsburgh Classifier System APCS. These algorithms were tested against baseline agents as well as the Online Evolutionary PlanningOEP algorithm. In a round-robin comparison among the agents, all LCS agents outperformed the baselines and OEP. APCS is the most effective game-playing agent while producing the most explainable output. ZCS and XCS outperformed the baselines and produced interpretable rule-sets. The results highlight the ability for symbolic methods to learn a complex wargame, outperform non-symbolic competitors, and provide replicable instructions.

Descriptors:

Subject Categories:

• Cybernetics
• Operations Research
• Military Operations, Strategy and Tactics

Distribution Statement:

[A, Approved For Public Release]