English

Regularized Risk Minimization by Nesterov's Accelerated Gradient Methods: Algorithmic Extensions and Empirical Studies

Machine Learning 2010-11-03 v1

Abstract

Nesterov's accelerated gradient methods (AGM) have been successfully applied in many machine learning areas. However, their empirical performance on training max-margin models has been inferior to existing specialized solvers. In this paper, we first extend AGM to strongly convex and composite objective functions with Bregman style prox-functions. Our unifying framework covers both the \infty-memory and 1-memory styles of AGM, tunes the Lipschiz constant adaptively, and bounds the duality gap. Then we demonstrate various ways to apply this framework of methods to a wide range of machine learning problems. Emphasis will be given on their rate of convergence and how to efficiently compute the gradient and optimize the models. The experimental results show that with our extensions AGM outperforms state-of-the-art solvers on max-margin models.

Keywords

Cite

@article{arxiv.1011.0472,
  title  = {Regularized Risk Minimization by Nesterov's Accelerated Gradient Methods: Algorithmic Extensions and Empirical Studies},
  author = {Xinhua Zhang and Ankan Saha and S. V. N. Vishwanathan},
  journal= {arXiv preprint arXiv:1011.0472},
  year   = {2010}
}

Comments

28 pages. Supplementary material for NIPS 2010 paper "Lower Bounds on Rate of Convergence of Cutting Plane Methods" by the same authors

R2 v1 2026-06-21T16:37:25.712Z