paper-with-me

홈 › Papers

Faster Acceleration for Steepest Descent

2024-09-28 · Cedar Site Bai, Brian Bullins

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.

📄 PDF Abstract BibTeX arXiv:2409.19200

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Approximate Steepest Coordinate Descent

2017-06-26 · ICML 2017 8 · Sebastian U. Stich, Anant Raj, Martin Jaggi

We propose a new selection rule for the coordinate selection in coordinate descent methods for huge-scale optimization. The efficiency of this novel scheme is provably better than the efficiency of uniformly random selec…

Computational Efficiencyregression

Faster gradient descent and the efficient recovery of images

2013-08-12 · Hui Huang, Uri Ascher

Much recent attention has been devoted to gradient descent algorithms where the steepest descent step size is replaced by a similar one from a previous iteration or gets updated only once every second step, thus forming …

DeblurringDenoisingImage Deblurring

Provable Benefit of Sign Descent: A Minimal Model Under Heavy-Tailed Class Imbalance

2025-11-30 · Robin Yadav, Shuo Xie, Tianhao Wang, Zhiyuan Li arxiv

Adaptive optimization methods (such as Adam) play a major role in LLM pretraining, significantly outperforming Gradient Descent (GD). Recent studies have proposed new smoothness assumptions on the loss function to explai…

Language Modelling

Interpreting Adaptive Gradient Methods by Parameter Scaling for Learning-Rate-Free Optimization

2024-01-06 · Min-Kook Suh, Seung-Woo Seo

We address the challenge of estimating the learning rate for adaptive gradient methods used in training deep neural networks. While several learning-rate-free approaches have been proposed, they are typically tailored fo…

Analyzing and Improving Greedy 2-Coordinate Updates for Equality-Constrained Optimization via Steepest Descent in the 1-Norm

2023-07-03 · Amrutha Varshini Ramesh, Aaron Mishkin, Mark Schmidt, Yihan Zhou 외

We consider minimizing a smooth function subject to a summation constraint over its variables. By exploiting a connection between the greedy 2-coordinate update for this problem and equality-constrained steepest descent …