Agent-mediated message passing for constrained environments

The majority of Multi-Agent Reinforcement Learning (MARL) literature equates the cooperation of self-interested agents in mixed environments to the problem of social welfare maximization, allowing agents to arbitrarily share rewards and private ...

We adopt the decision-theoretic principle of expected utility maximization as a paradigm for designing autonomous rational agents, and present a framework that uses this paradigm to determine the choice of coordinated action. We endow an agent with a ...

This work reflects continued research into "temporally autonomous" multi-agent interaction. Many traditional approaches to modeling multi-agent systems involve synchronizing all agent activity in simulated environments to a single "universal'' clock. In ...