21
57

Distributed Online Optimization with Long-Term Constraints

Abstract

We consider distributed online convex optimization problems, where the distributed system consists of various computing units connected through a time-varying communication graph. In each time step, each computing unit selects a constrained vector, experiences a loss equal to an arbitrary convex function evaluated at this vector, and may communicate to its neighbors in the graph. The objective is to minimize the system-wide loss accumulated over time. We propose a decentralized algorithm with regret and cumulative constraint violation in O(Tmax{c,1c})\mathcal{O}(T^{\max\{c,1-c\} }) and O(T1c/2)\mathcal{O}(T^{1-c/2}), respectively, for any c(0,1)c\in (0,1), where TT is the time horizon. When the loss functions are strongly convex, we establish improved regret and constraint violation upper bounds in O(log(T))\mathcal{O}(\log(T)) and O(Tlog(T))\mathcal{O}(\sqrt{T\log(T)}). These regret scalings match those obtained by state-of-the-art algorithms and fundamental limits in the corresponding centralized online optimization problem (for both convex and strongly convex loss functions). In the case of bandit feedback, the proposed algorithms achieve a regret and constraint violation in O(Tmax{c,1c/3})\mathcal{O}(T^{\max\{c,1-c/3 \} }) and O(T1c/2)\mathcal{O}(T^{1-c/2}) for any c(0,1)c\in (0,1). We numerically illustrate the performance of our algorithms for the particular case of distributed online regularized linear regression problems.

View on arXiv
Comments on this paper