Factored Beliefs for Machine Agents in Decentralized Partially Observable Markov Decision Processes

Authors

  • Joshua Lapso Air Force Institute of Technology
  • Gilbert Peterson Air Force Institute of Technology

DOI:

https://doi.org/10.32473/flairs.v35i.130671

Keywords:

Shared Mental Model, Transactive Memory System, Decentralized Partially Observable Markov Decision Process, Dynamic Bayesian Network, Multi-agent System

Abstract

A shared mental model (SMM) is a foundational structure in high performing, task-oriented teams and aid humans in determining their teammate's goals and intentions. Higher levels of mental alignment between teammates can reduce the direct dialogue required for team success. For decision-making teams, a transactive memory system (TMS) offers team members a map of specialized knowledge, indicating source of knowledge and the source's credibility. SMM and TMS formulations aid human-agent team performance in their intended team types. However, neither improve team performance with a project team--one that requires both behavioral and knowledge integration. We present a hybrid cognitive model (HCM) for machine agents that subsumes the integrated portions of a team's transactive memory in an SMM. The unified structure of the HCM enables contextual switches during execution for machine agents, over the two cognitive formulations with comparable computational complexity of a single cognitive model. Results in a multi-agent project environment demonstrates how the HCM provides machine agents with a generalizable cognitive structure that is able to maintain fully factored belief states with minimal inter-agent communication.

Author Biographies

Joshua Lapso, Air Force Institute of Technology

PhD Candidate (Department of Electrical and Computer Engineering)

Gilbert Peterson, Air Force Institute of Technology

Professor of Computer Science (Department of Electrical and Computer Engineering)

Downloads

Published

04-05-2022

How to Cite

Lapso, J., & Peterson, G. (2022). Factored Beliefs for Machine Agents in Decentralized Partially Observable Markov Decision Processes. The International FLAIRS Conference Proceedings, 35. https://doi.org/10.32473/flairs.v35i.130671

Issue

Section

Special Track: Uncertain Reasoning