Convergence of Langevin-simulated annealing algorithms with multiplicative noise. II: Total variation

DOI10.1515/mcma-2023-2009arXiv2205.15039OpenAlexW4310891583MaRDI QIDQ6073725

Pierre Bras, Gilles Pagès

Publication date: 18 September 2023

Published in: Monte Carlo Methods and Applications (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2205.15039

zbMATH Keywords

stochastic optimizationneural networkssimulated annealingLangevin equation

Mathematics Subject Classification ID

Stochastic approximation (62L20) Computational methods for stochastic equations (aspects of stochastic analysis) (60H35) Numerical solutions to stochastic differential and integral equations (65C30)

Cites Work

• Unnamed Item
• Unnamed Item
• Recursive computation of the invariant distribution of a diffusion
• Laplace's method revisited: Weak convergence of probability measures
• Total variation distance between two diffusions in small time with unbounded drift: application to the Euler-Maruyama scheme
• Density and gradient estimates for non degenerate Brownian SDEs with unbounded measurable drift
• High-dimensional Bayesian inference via the unadjusted Langevin algorithm
• Nonasymptotic convergence analysis for the unadjusted Langevin algorithm
• A representation formula for transition probability densities of diffusions and applications
• Convergence rates of Gibbs measures with degenerate minimum
• Theoretical Guarantees for Approximate Sampling from Smooth and Log-Concave Densities
• Optimal Transport
• Unadjusted Langevin algorithm with multiplicative noise: total variation and Wasserstein bounds

This page was built for publication: Convergence of Langevin-simulated annealing algorithms with multiplicative noise. II: Total variation