OCLGMLJun 22, 2020

Gradient-Variation Bound for Online Convex Optimization with Constraints

arXiv:2006.12455v36 citations
Originality Highly original
AI Analysis

This addresses computational efficiency and performance in constrained online optimization for machine learning and operations research, offering a non-incremental improvement with broader applicability in general normed spaces.

The paper tackles online convex optimization with constraints by proposing a novel primal-dual mirror-prox algorithm that achieves an instance-dependent regret bound of O(√V_*(T)) and O(1) constraint violation, improving over previous O(√T) and O(T^{2/3}) bounds.

We study online convex optimization with constraints consisting of multiple functional constraints and a relatively simple constraint set, such as a Euclidean ball. As enforcing the constraints at each time step through projections is computationally challenging in general, we allow decisions to violate the functional constraints but aim to achieve a low regret and cumulative violation of the constraints over a horizon of $T$ time steps. First-order methods achieve an $\mathcal{O}(\sqrt{T})$ regret and an $\mathcal{O}(1)$ constraint violation, which is the best-known bound under the Slater's condition, but do not take into account the structural information of the problem. Furthermore, the existing algorithms and analysis are limited to Euclidean space. In this paper, we provide an \emph{instance-dependent} bound for online convex optimization with complex constraints obtained by a novel online primal-dual mirror-prox algorithm. Our instance-dependent regret is quantified by the total gradient variation $V_*(T)$ in the sequence of loss functions. The proposed algorithm works in \emph{general} normed spaces and simultaneously achieves an $\mathcal{O}(\sqrt{V_*(T)})$ regret and an $\mathcal{O}(1)$ constraint violation, which is never worse than the best-known $( \mathcal{O}(\sqrt{T}), \mathcal{O}(1) )$ result and improves over previous works that applied mirror-prox-type algorithms for this problem achieving $\mathcal{O}(T^{2/3})$ regret and constraint violation. Finally, our algorithm is computationally efficient, as it only performs mirror descent steps in each iteration instead of solving a general Lagrangian minimization problem.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes