Scalable Frank-Wolfe on Generalized Self-concordant Functions via Simple Steps

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Carderera, Alejandro, Besançon, Mathieu, Pokutta, Sebastian
Format: Preprint
Published: 2021
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911829041610752
author Carderera, Alejandro
Besançon, Mathieu
Pokutta, Sebastian
author_facet Carderera, Alejandro
Besançon, Mathieu
Pokutta, Sebastian
contents Generalized self-concordance is a key property present in the objective function of many important learning problems. We establish the convergence rate of a simple Frank-Wolfe variant that uses the open-loop step size strategy $γ_t = 2/(t+2)$, obtaining a $\mathcal{O}(1/t)$ convergence rate for this class of functions in terms of primal gap and Frank-Wolfe gap, where $t$ is the iteration count. This avoids the use of second-order information or the need to estimate local smoothness parameters of previous work. We also show improved convergence rates for various common cases, e.g., when the feasible region under consideration is uniformly convex or polyhedral.
format Preprint
id arxiv_https___arxiv_org_abs_2105_13913
institution arXiv
publishDate 2021
record_format arxiv
spellingShingle Scalable Frank-Wolfe on Generalized Self-concordant Functions via Simple Steps
Carderera, Alejandro
Besançon, Mathieu
Pokutta, Sebastian
Optimization and Control
Machine Learning
Generalized self-concordance is a key property present in the objective function of many important learning problems. We establish the convergence rate of a simple Frank-Wolfe variant that uses the open-loop step size strategy $γ_t = 2/(t+2)$, obtaining a $\mathcal{O}(1/t)$ convergence rate for this class of functions in terms of primal gap and Frank-Wolfe gap, where $t$ is the iteration count. This avoids the use of second-order information or the need to estimate local smoothness parameters of previous work. We also show improved convergence rates for various common cases, e.g., when the feasible region under consideration is uniformly convex or polyhedral.
title Scalable Frank-Wolfe on Generalized Self-concordant Functions via Simple Steps
topic Optimization and Control
Machine Learning
url https://arxiv.org/abs/2105.13913