Partially Observable Stochastic Games with Neural Perception Mechanisms
This work addresses the challenge of integrating neural perception into multi-agent systems for domains like autonomous driving and robotics, though it appears incremental as it builds on existing stochastic game models.
The paper tackles the problem of multi-agent decision-making under partial observability with neural perception by introducing neuro-symbolic partially-observable stochastic games (NS-POSGs) and a one-sided NS-HSVI method for approximate solution, demonstrating practical applicability in pedestrian-vehicle and pursuit-evasion scenarios.
Stochastic games are a well established model for multi-agent sequential decision making under uncertainty. In practical applications, though, agents often have only partial observability of their environment. Furthermore, agents increasingly perceive their environment using data-driven approaches such as neural networks trained on continuous data. We propose the model of neuro-symbolic partially-observable stochastic games (NS-POSGs), a variant of continuous-space concurrent stochastic games that explicitly incorporates neural perception mechanisms. We focus on a one-sided setting with a partially-informed agent using discrete, data-driven observations and another, fully-informed agent. We present a new method, called one-sided NS-HSVI, for approximate solution of one-sided NS-POSGs, which exploits the piecewise constant structure of the model. Using neural network pre-image analysis to construct finite polyhedral representations and particle-based representations for beliefs, we implement our approach and illustrate its practical applicability to the analysis of pedestrian-vehicle and pursuit-evasion scenarios.