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

Bras Pierre () and Pagès Gilles ()
Additional contact information
Bras Pierre: Laboratoire de Probabilités, Statistiques et Modélisation, Sorbonne Université, Paris, France
Pagès Gilles: Laboratoire de Probabilités, Statistiques et Modélisation, Sorbonne Université, Paris, France

Monte Carlo Methods and Applications, 2023, vol. 29, issue 3, 203-219

Abstract: We study the convergence of Langevin-simulated annealing type algorithms with multiplicative noise, i.e. for V : R d → R V\colon\mathbb{R}^{d}\to\mathbb{R} a potential function to minimize, we consider the stochastic differential equation d ⁢ Y t = − σ ⁢ σ ⊤ ⁢ ∇ V ⁢ ( Y t ) ⁢ d ⁢ t + a ⁢ ( t ) ⁢ σ ⁢ ( Y t ) ⁢ d ⁢ W t + a ⁢ ( t ) 2 ⁢ Υ ⁢ ( Y t ) ⁢ d ⁢ t dY_{t}=-\sigma\sigma^{\top}\nabla V(Y_{t})\,dt+a(t)\sigma(Y_{t})\,dW_{t}+a(t)^{2}\Upsilon(Y_{t})\,dt , where ( W t ) (W_{t}) is a Brownian motion, σ : R d → M d ⁢ ( R ) \sigma\colon\mathbb{R}^{d}\to\mathcal{M}_{d}(\mathbb{R}) is an adaptive (multiplicative) noise, a : R + → R + a\colon\mathbb{R}^{+}\to\mathbb{R}^{+} is a function decreasing to 0 and where Υ is a correction term. Allowing 𝜎 to depend on the position brings faster convergence in comparison with the classical Langevin equation d ⁢ Y t = − ∇ V ⁢ ( Y t ) ⁢ d ⁢ t + σ ⁢ d ⁢ W t dY_{t}=-\nabla V(Y_{t})\,dt+\sigma\,dW_{t} . In a previous paper, we established the convergence in L 1 L^{1} -Wasserstein distance of Y t Y_{t} and of its associated Euler scheme Y ¯ t \bar{Y}_{t} to argmin ⁡ ( V ) \operatorname{argmin}(V) with the classical schedule a ⁢ ( t ) = A ⁢ log − 1 / 2 ⁡ ( t ) a(t)=A\log^{-1/2}(t) . In the present paper, we prove the convergence in total variation distance. The total variation case appears more demanding to deal with and requires regularization lemmas.

Keywords: Stochastic optimization; Langevin equation; simulated annealing; neural networks (search for similar items in EconPapers)
Date: 2023
References: Add references at CitEc
Citations:

Downloads: (external link)
https://doi.org/10.1515/mcma-2023-2009 (text/html)
For access to full text, subscription to the journal or payment for the individual article is required.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:bpj:mcmeap:v:29:y:2023:i:3:p:203-219:n:1

Ordering information: This journal article can be ordered from
https://www.degruyter.com/journal/key/mcma/html

DOI: 10.1515/mcma-2023-2009

Access Statistics for this article

Monte Carlo Methods and Applications is currently edited by Karl K. Sabelfeld

More articles in Monte Carlo Methods and Applications from De Gruyter
Bibliographic data for series maintained by Peter Golla ().