Zeroth-order non-convex learning via hierarchical dual averaging - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2021

Zeroth-order non-convex learning via hierarchical dual averaging

Abstract

We propose a hierarchical version of dual averaging for zeroth-order online non-convex optimization-i.e., learning processes where, at each stage, the optimizer is facing an unknown non-convex loss function and only receives the incurred loss as feedback. The proposed class of policies relies on the construction of an online model that aggregates loss information as it arrives, and it consists of two principal components: (a) a regularizer adapted to the Fisher information metric (as opposed to the metric norm of the ambient space); and (b) a principled exploration of the problem's state space based on an adapted hierarchical schedule. This construction enables sharper control of the model's bias and variance, and allows us to derive tight bounds for both the learner's static and dynamic regret-i.e., the regret incurred against the best dynamic policy in hindsight over the horizon of play.
Fichier principal
Vignette du fichier
2021-ICML-HDA.pdf (1.22 Mo) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03341912 , version 1 (13-09-2021)

Identifiers

  • HAL Id : hal-03341912 , version 1

Cite

Amélie Héliou, Matthieu Martin, Thibaud Rahier, Panayotis Mertikopoulos. Zeroth-order non-convex learning via hierarchical dual averaging. ICML 2021 - 38th International Conference on Machine Learning, Jul 2021, Vienna, Austria. pp.1-34. ⟨hal-03341912⟩
63 View
68 Download

Share

Gmail Mastodon Facebook X LinkedIn More