A Deep Look into Trust and Mutual Understanding in Multi-Agent Cooperative Game Through Explainable Reinforcement Learning
Abstract
Establishing trust in autonomous systems trained by machine learning is one of the important steps towards fostering their acceptance. This requires those systems to provide explanations for their actions in a manner that is intuitive and understandable to humans. In a multi-agent cooperative environment where numerous agents collaborate to achieve a common mission, the trust existing among these agents, a concept inherently familiar to humans, significantly influences the success of the mission. Agent-to-agent (A2A) trust will become increasingly important as disparate companies, agencies, and other organizations deploy heterogenous autonomous systems that operate in a shared environment.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Feb 05, 2025
- Source ID
- FA95502410078
Entities
People
- Qinru Qiu
Organizations
- Air Force Office of Scientific Research
- Syracuse University
- United States Air Force