stat.MLJul 21, 2026

The Price of Hidden Curvature: An Ω~(d5/4T)\widetildeΩ (d^{5/4} \sqrt{T}) Lower Bound for Bandit Convex Optimization

Authors: Nived Rajaraman

Organizations: Microsoft Research

Abstract

We establish a Ω~(d5/4T)\widetildeΩ(d^{5/4}\sqrt T) lower bound on the minimax expected regret of stochastic bandit convex optimization of 11-Lipschitz functions on the Euclidean ball. This presents the first nontrivial regret lower bound that grows faster than dTd\sqrt{T} for this problem, establishing that stochastic bandit convex optimization is fundamentally harder than linear bandits. The hard class of convex functions we construct takes the following form in dimension 2d2d: for an action a=(a1,a2)B22da = (a^1,a^2) \in \mathbb{B}^{2d}_2, each function is the scaled soft maximum of a "tube", r1Wa1r8εa22r^{-1} \| W^\star a^1 - \frac{r}{8\varepsilon} a^2 \|_2 (hyperparameterized by ε,r\varepsilon,r), and a squared distance function, 12a1u2212u22\frac12 \| a^1 - u^\star \|_2^2 - \frac12 \| u^\star \|_2^2. Here, WRd×dW^\star \in \mathbb{R}^{d \times d} is an unknown linear transformation, and uRdu^\star \in \mathbb{R}^{d} is an unknown vector which must be learned to minimize the function. Observations are informative about uu^\star only when the learner's action lies near the tube determined by WW^\star, satisfying a28εrWa1a^2 \approx \frac{8\varepsilon}{r} W^\star a^1: thus the learner must either find this tube without knowing WW^\star, or spend observations learning useful directions of WW^\star. Formally, our regret analysis exploits this tradeoff by bounding the posterior spread of Fisher information matrices obtained under an adaptive sequence of actions. Together, these ingredients give a sample complexity lower bound of Ω~(d5/2/ε2)\widetildeΩ(d^{5/2}/\varepsilon^2) to find an ε\varepsilon-optimal action, which translates to an Ω~(d5/4T)\widetildeΩ (d^{5/4} \sqrt{T}) regret lower bound. We also extend this lower bound to the unconstrained setting where the action space is Rd\mathbb{R}^d.

Explore similar work

CardsList
  1. Small Gradient Norm Regret for Online Convex Optimization

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