Faster Acceleration for Steepest Descent
This addresses a long-standing barrier in accelerated non-Euclidean steepest descent for optimization researchers, though it appears incremental as it builds on prior advances in dimension dependence.
The paper tackles the problem of accelerating first-order methods for convex optimization under non-Euclidean smoothness, specifically for ℓ_p norm smooth functions, achieving an iteration complexity improvement of up to O(d^{1-2/p}) in terms of first-order oracle calls.
Recent advances (Sherman, 2017; Sidford and Tian, 2018; Cohen et al., 2021) have overcome the fundamental barrier of dimension dependence in the iteration complexity of solving $\ell_\infty$ regression with first-order methods. Yet it remains unclear to what extent such acceleration can be achieved for general $\ell_p$ smooth functions. In this paper, we propose a new accelerated first-order method for convex optimization under non-Euclidean smoothness assumptions. In contrast to standard acceleration techniques, our approach uses primal-dual iterate sequences taken with respect to $\textit{differing}$ norms, which are then coupled using an $\textit{implicitly}$ determined interpolation parameter. For $\ell_p$ norm smooth problems in $d$ dimensions, our method provides an iteration complexity improvement of up to $O(d^{1-\frac{2}{p}})$ in terms of calls to a first-order oracle, thereby allowing us to circumvent long-standing barriers in accelerated non-Euclidean steepest descent.