An Optimistic Algorithm for Online Convex Optimization with Adversarial Constraints

Abstract

We study Online Convex Optimization (OCO) with adversarial constraints, where an online algorithm must make sequential decisions to minimize both convex loss functions and cumulative constraint violations. We focus on a setting where the algorithm has access to predictions of the loss and constraint functions. Our results show that we can improve the current best bounds of O(T) regret and O(T) cumulative constraint violations to O(E_T(f)) and O(E_T(g^+)) , respectively, where E_T(f) and E_T(g^+) represent the cumulative prediction errors of the loss and constraint functions. In the worst case, where E_T(f) = O(T) and E_T(g^+) = O(T) (assuming bounded gradients of the loss and constraint functions), our rates match the prior O(T) results. However, when the loss and constraint predictions are accurate, our approach yields significantly smaller regret and cumulative constraint violations. Finally, we apply this to the setting of adversarial contextual bandits with sequential risk constraints, obtaining optimistic bounds O (E_T(f) T^1/3) regret and O(E_T(g^+) T^1/3) constraints violation, yielding better performance than existing results when prediction quality is sufficiently high.

Tasks

Reproductions