Expected flow networks in stochastic environments and two-player zero-sum games

Physics-informed adversarial networks for stochastic second-order mean-field games of autonomous traffic

Inspired by the highly complex and nonlinear traffic flow phenomenon, we propose a second-order continuum model for autonomous vehicles (AVs) by considering them as rational and utility-optimizing agents. Further, by accounting for stochastic fluctuations in driving dynamics, a non-cooperative differential game probl...

Unifying Generative Models with Path Integrals

We formulate generative modeling as a path integral in which flow-based, diffusion-based, variational, and adversarial models arise as different evaluation principles for a single master action. Its Martin-Siggia-Rose-Janssen-de~Dominicis (MSRJD) form separates free from interacting probability flows and opens them to...

Constant Swap Regret in General-Sum Games via Optimistic Transition Matrices

RLVR landscapes for iterated multiplications can be benign: Insights from spin-glass theory

This work maps entropy-regularized RLVR over myopic tabular policies onto an energy-based (spin-glass) model over deterministic policies, and finds that a transformer trained from scratch, using only last-token rewards, successfully learns an algorithmic chain of thought for iterated non-Abelian group multiplications.

Reward-guided Fine-Tuning of One-Step Generative Models via Wasserstein Gradient Flow

This work considers one-step generators from an optimal transport view, investigating Wasserstein Gradient Flow (WGF) for modeling smooth and controlled distributional evolution in probability space, and proposes a novel reward-guided fine-tuning of a one-step generative model via WGF.

Constant Swap Regret in General-Sum Games via Two-Scale Higher-Order Optimism

We give deterministic and uncoupled learning dynamics for finite multiplayer general-sum games under full-information feedback that achieve constant individual swap regret in self-play, independent of the horizon $T$. With $n$ players and at most $m$ actions each, every player's individual swap regret is $O(\sqrt n\,m\...