Analogical Learning in Tactical Decision Games

TL;DR

This paper explores analogical learning methods for tactical decision games, demonstrating how incremental learning and mapping constraints can improve AI performance in complex military scenarios.

Contribution

It introduces novel analogical learning techniques tailored for the complex, open-ended domain of Tactical Decision Games, including partition constraints and incremental remapping.

Findings

Performance improved with example accumulation

Analogical mapping constraints enhance robustness

Weak domain theory still yields effective learning

Abstract

Tactical Decision Games (TDGs) are military conflict scenarios presented both textually and graphically on a map. These scenarios provide a challenging domain for machine learning because they are open-ended, highly structured, and typically contain many details of varying relevance. We have developed a problem-solving component of an interactive companion system that proposes military tasks to solve TDG scenarios using a combination of analogical retrieval, mapping, and constraint propagation. We use this problem-solving component to explore analogical learning. In this paper, we describe the problems encountered in learning for this domain, and the methods we have developed to address these, such as partition constraints on analogical mapping correspondences and the use of incremental remapping to improve robustness. We present the results of learning experiments that show improvement in performance through the simple accumulation of examples, despite a weak domain theory.