A Second-order Bound with Excess Losses - Archive ouverte HAL Access content directly
Preprints, Working Papers, ... Year : 2014

A Second-order Bound with Excess Losses

Abstract

We study online aggregation of the predictions of experts, and first show new second-order regret bounds in the standard setting, which are obtained via a version of the Prod algorithm (and also a version of the polynomially weighted average algorithm) with multiple learning rates. These bounds are in terms of excess losses, the differences between the instantaneous losses suffered by the algorithm and the ones of a given expert. We then demonstrate the interest of these bounds in the context of experts that report their confidences as a number in the interval [0,1] using a generic reduction to the standard setting. We conclude by two other applications in the standard setting, which improve the known bounds in case of small excess losses and show a bounded regret against i.i.d. sequences of losses.
Fichier principal
Vignette du fichier
sleeping-experts.pdf (279.61 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-00943665 , version 1 (08-02-2014)

Identifiers

Cite

Pierre Gaillard, Gilles Stoltz, Tim van Erven. A Second-order Bound with Excess Losses. 2014. ⟨hal-00943665⟩
445 View
257 Download

Altmetric

Share

Gmail Facebook X LinkedIn More