Abstract
Strategic reasoning is essential to ensure stable multi-agent coordination in complex environments, as it enables synthesis of optimal (or near-optimal) agent strategies and equilibria that guarantee expected outcomes, even in adversarial scenarios. Partially-observable stochastic games (POSGs) are a natural model for real-world settings involving multiple agents, uncertainty and partial information, but lack practical algorithms for computing or approximating optimal values and strategies. This talk will discuss progress with developing a model class and algorithms for one-sided POSGs with neural perception mechanisms used to observe their continuous environment. Building on the fact is that ReLU neural-network classifiers induce a finite decomposition of the continuous environment into polyhedral for each classification label, an abstract piecewise constant representation of value functions is derived, and a point-based solution method that computes a lower and upper bound on the value function from a given belief to compute an (approximately) optimal strategy.