General parallel optimization without a metric - Archive ouverte HAL Access content directly
Conference Papers Year : 2019

General parallel optimization without a metric

Xuedong Shang
Michal Valko

Abstract

Hierarchical bandits are an approach for global optimization of extremely irregular functions. This paper provides new elements regarding POO, an adaptive meta-algorithm that does not require the knowledge of local smoothness of the target function. We first highlight the fact that the subroutine algorithm used in POO should have a small regret under the assumption of local smoothness with respect to the chosen partitioning, which is unknown if it is satisfied by the standard subroutine HOO. In this work, we establish such regret guarantee for HCT, which is another hierarchical optimistic optimization algorithm that needs to know the smoothness. This confirms the validity of POO. We show that POO can be used with HCT as a subroutine with a regret upper bound that matches the one of best-known algorithms using the knowledge of smoothness up to a √ log n factor. On top of that, we propose a general wrapper, called GPO, that can cope with algorithms that only have simple regret guarantees. Finally, we complement our findings with experiments on difficult functions.
Fichier principal
Vignette du fichier
shang2019general.pdf (5.04 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-02047225 , version 1 (24-02-2019)
hal-02047225 , version 2 (05-03-2019)

Identifiers

  • HAL Id : hal-02047225 , version 2

Cite

Xuedong Shang, Emilie Kaufmann, Michal Valko. General parallel optimization without a metric. Algorithmic Learning Theory, 2019, Chicago, United States. ⟨hal-02047225v2⟩
267 View
173 Download

Share

Gmail Facebook Twitter LinkedIn More