A connection between Tempering and Entropic Mirror Descent

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Chopin, Nicolas, Crucinio, Francesca R., Korba, Anna
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914834802540544
author Chopin, Nicolas
Crucinio, Francesca R.
Korba, Anna
author_facet Chopin, Nicolas
Crucinio, Francesca R.
Korba, Anna
contents This paper explores the connections between tempering (for Sequential Monte Carlo; SMC) and entropic mirror descent to sample from a target probability distribution whose unnormalized density is known. We establish that tempering SMC corresponds to entropic mirror descent applied to the reverse Kullback-Leibler (KL) divergence and obtain convergence rates for the tempering iterates. Our result motivates the tempering iterates from an optimization point of view, showing that tempering can be seen as a descent scheme of the KL divergence with respect to the Fisher-Rao geometry, in contrast to Langevin dynamics that perform descent of the KL with respect to the Wasserstein-2 geometry. We exploit the connection between tempering and mirror descent iterates to justify common practices in SMC and derive adaptive tempering rules that improve over other alternative benchmarks in the literature.
format Preprint
id arxiv_https___arxiv_org_abs_2310_11914
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle A connection between Tempering and Entropic Mirror Descent
Chopin, Nicolas
Crucinio, Francesca R.
Korba, Anna
Computation
Optimization and Control
Statistics Theory
Machine Learning
This paper explores the connections between tempering (for Sequential Monte Carlo; SMC) and entropic mirror descent to sample from a target probability distribution whose unnormalized density is known. We establish that tempering SMC corresponds to entropic mirror descent applied to the reverse Kullback-Leibler (KL) divergence and obtain convergence rates for the tempering iterates. Our result motivates the tempering iterates from an optimization point of view, showing that tempering can be seen as a descent scheme of the KL divergence with respect to the Fisher-Rao geometry, in contrast to Langevin dynamics that perform descent of the KL with respect to the Wasserstein-2 geometry. We exploit the connection between tempering and mirror descent iterates to justify common practices in SMC and derive adaptive tempering rules that improve over other alternative benchmarks in the literature.
title A connection between Tempering and Entropic Mirror Descent
topic Computation
Optimization and Control
Statistics Theory
Machine Learning
url https://arxiv.org/abs/2310.11914