cs.LGMay 13, 2026

Polyhedral Instability Governs Regret in Online Learning

Authors: Yuetai LiFengqing JiangYichen FengKaiyuan ZhengLuyao NiuBhaskar RamasubramanianBasel AlomairLinda Bushnell+1 more

Organizations: ♣University of Washington · ♢Western Washington University · ♠King Abdulaziz City for Science and Technology ♡HUMAIN

Abstract

Many online decision problems over combinatorial actions are addressed via convex relaxations, leading to online convex optimization with piecewise linear objectives and induced polyhedral structure. We show that regret in such problems is governed by \emph{polyhedral instability}: the number of changes of the active region. Under full information feedback and fixed partition assumptions, if RST\mathrm{RS}_T denotes the number of region switches and VmaxV_{\max} the maximum number of vertices per region, we prove \RegretT=Θ((1+RST)TlogVmax)\Regret_T= Θ(\sqrt{(1+\mathrm{RS}_T)\,T\,\log V_{\max}}) interpolating between experts-like and dimension-dependent OCO rates. For online submodular--concave games under Lovász convexification, this reduces to the permutation-switch count SCT\mathrm{SC}_T, yielding the matching rate \RegretT=Θ((1+SCT)Tlogn)\Regret_T= Θ(\sqrt{(1+\mathrm{SC}_T)\,T\,\log n}). Experiments on synthetic and real combinatorial problems (shortest path, influence maximization) validate the predicted scaling and indicate that low-instability regimes can arise in practice without explicit enumeration of actions.

Explore similar work

CardsList
  1. Small Gradient Norm Regret for Online Convex Optimization

    Jan 20, 2026Wenzhi Gao, Chang He, Madeleine Udell