A Generic Approach for Escaping Saddle points

Abstract : A central challenge to using first-order methods for optimizing nonconvex problems is the presence of saddle points. First-order methods often get stuck at saddle points, greatly deteriorating their performance. Typically, to escape from saddles one has to use second-order methods. However, most works on second-order methods rely extensively on expensive Hessian-based computations, making them impractical in large-scale settings. To tackle this challenge, we introduce a generic framework that minimizes Hessian based computations while at the same time provably converging to second-order critical points. Our framework carefully alternates between a first-order and a second-order subroutine, using the latter only close to saddle points, and yields convergence results competitive to the state-of-the-art. Empirical results suggest that our strategy also enjoys a good practical performance.
Type de document :
Communication dans un congrès
AISTATS 2018 - 22nd International Conference on Artificial Intelligence and Statistics, Apr 2018, Canary Islands, Spain
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01652150
Contributeur : Francis Bach <>
Soumis le : jeudi 30 novembre 2017 - 07:56:23
Dernière modification le : mercredi 30 janvier 2019 - 11:07:45

Lien texte intégral

Identifiants

  • HAL Id : hal-01652150, version 1
  • ARXIV : 1709.01434

Collections

Citation

Sashank J Reddi, Manzil Zaheer, Suvrit Sra, Barnabas Poczos, Francis Bach, et al.. A Generic Approach for Escaping Saddle points. AISTATS 2018 - 22nd International Conference on Artificial Intelligence and Statistics, Apr 2018, Canary Islands, Spain. 〈hal-01652150〉

Partager

Métriques

Consultations de la notice

178