Riemannian Manifold Learning for Stackelberg Games with Neural Flow Representations

TL;DR

This paper introduces a novel manifold learning framework using neural flows for online Stackelberg game learning, enabling efficient equilibrium computation with theoretical regret bounds and empirical validation in cybersecurity and supply chains.

Contribution

It proposes a new approach combining Riemannian manifold learning with neural normalizing flows for Stackelberg games, providing theoretical regret guarantees and demonstrating practical effectiveness.

Findings

Effective regret minimization on the learned manifold.

Superior performance over standard baselines.

Applicability to cybersecurity and supply chain domains.

Abstract

We present a novel framework for online learning in Stackelberg general-sum games, where two agents, the leader and follower, engage in sequential turn-based interactions. At the core of this approach is a learned diffeomorphism that maps the joint action space to a smooth spherical Riemannian manifold, referred to as the Stackelberg manifold. This mapping, facilitated by neural normalizing flows, ensures the formation of tractable isoplanar subspaces, enabling efficient techniques for online learning. Leveraging the linearity of the agents' reward functions on the Stackelberg manifold, our construct allows the application of linear bandit algorithms. We then provide a rigorous theoretical basis for regret minimization on the learned manifold and establish bounds on the simple regret for learning Stackelberg equilibrium. This integration of manifold learning into game theory uncovers a previously unrecognized potential for neural normalizing flows as an effective tool for multi-agent learning. We present empirical results demonstrating the effectiveness of our approach compared to standard baselines, with applications spanning domains such as cybersecurity and economic supply chain optimization.