Submitted:
16 November 2026
Posted:
18 November 2026
You are already at the latest version
Abstract
We present two methods for solving non-cooperative differential games as variational inference problems. We extend Stein Variational Model Predictive Control and a Cross-Entropy method used to solve stochastic optimal control problems to the adversarial setting so that they can be used to solve differential games of mixed strategies. We demonstrate the methods on a perimeter defense problem where two groups of robots interact with conflicting objectives
Keywords:
Model Predictive Control
; Monte Carlo simulation
; stochastic model
; variational Inference
; cross entropy
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.