Saved in:
Bibliographic Details
Main Author: Zhou, Quan
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.00708
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909882255409152
author Zhou, Quan
author_facet Zhou, Quan
contents We study the theoretical complexity of simulated tempering for sampling from mixtures of log-concave components differing only by location shifts. The main result establishes the first polynomial-time guarantee for simulated tempering combined with the Metropolis-adjusted Langevin algorithm (MALA) with respect to the problem dimension $d$, maximum mode displacement $D$, and logarithmic accuracy $\log ε^{-1}$. The proof builds on a general state decomposition theorem for $s$-conductance, applied to an auxiliary Markov chain constructed on an augmented space. We also obtain an improved complexity estimate for simulated tempering combined with random-walk Metropolis. Our bounds assume an inverse-temperature ladder with smallest value $β_1 = O(D^{-2})$ and spacing $β_{i+1}/β_i = 1 + O( d^{-1/2} )$, both of which are shown to be asymptotically optimal up to logarithmic factors.
format Preprint
id arxiv_https___arxiv_org_abs_2511_00708
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Polynomial Mixing Times of Simulated Tempering for Mixture Targets by Conductance Decomposition
Zhou, Quan
Computation
Probability
Machine Learning
60J20, 65C05, 65C40, 68Q25
We study the theoretical complexity of simulated tempering for sampling from mixtures of log-concave components differing only by location shifts. The main result establishes the first polynomial-time guarantee for simulated tempering combined with the Metropolis-adjusted Langevin algorithm (MALA) with respect to the problem dimension $d$, maximum mode displacement $D$, and logarithmic accuracy $\log ε^{-1}$. The proof builds on a general state decomposition theorem for $s$-conductance, applied to an auxiliary Markov chain constructed on an augmented space. We also obtain an improved complexity estimate for simulated tempering combined with random-walk Metropolis. Our bounds assume an inverse-temperature ladder with smallest value $β_1 = O(D^{-2})$ and spacing $β_{i+1}/β_i = 1 + O( d^{-1/2} )$, both of which are shown to be asymptotically optimal up to logarithmic factors.
title Polynomial Mixing Times of Simulated Tempering for Mixture Targets by Conductance Decomposition
topic Computation
Probability
Machine Learning
60J20, 65C05, 65C40, 68Q25
url https://arxiv.org/abs/2511.00708