Towards A Theory-Of-Mind-Inspired Generic Decision-Making Framework

TL;DR

This paper introduces a theory-of-mind-inspired decision-making framework that enables agents to reason and act using multiple simulations of their environment, demonstrated through a game-playing application and compared to benchmarks.

Contribution

It presents a novel generic framework based on theory of mind for decision-making in complex environments, with a partial implementation and benchmark comparison.

Findings

The framework successfully played a game in the IJCAI2013 AIBirds contest.

It outperformed some baseline strategies in the benchmark.

The approach demonstrates the potential of simulation-based reasoning inspired by cognitive science.

Abstract

Simulation is widely used to make model-based predictions, but few approaches have attempted this technique in dynamic physical environments of medium to high complexity or in general contexts. After an introduction to the cognitive science concepts from which this work is inspired and the current development in the use of simulation as a decision-making technique, we propose a generic framework based on theory of mind, which allows an agent to reason and perform actions using multiple simulations of automatically created or externally inputted models of the perceived environment. A description of a partial implementation is given, which aims to solve a popular game within the IJCAI2013 AIBirds contest. Results of our approach are presented, in comparison with the competition benchmark. Finally, future developments regarding the framework are discussed.