A Mini-Batch Proximal Stochastic Recursive Gradient Algorithm with Diagonal Barzilai–Borwein Stepsize

Many machine learning problems can be formulated as minimizing the sum of a function and a non-smooth regularization term. Proximal stochastic gradient methods are popular for solving such composite optimization problems. We propose a mini-batch proximal stochastic recursive gradient algorithm SRG-D...

Full description

Bibliographic Details
Main Authors: Yu, T.T., Liu, X.W., Dai, Y.H., Sun, Jie
Format: Journal Article
Published: 2022
Online Access:http://hdl.handle.net/20.500.11937/91425
_version_ 1848765517044645888
author Yu, T.T.
Liu, X.W.
Dai, Y.H.
Sun, Jie
author_facet Yu, T.T.
Liu, X.W.
Dai, Y.H.
Sun, Jie
author_sort Yu, T.T.
building Curtin Institutional Repository
collection Online Access
description Many machine learning problems can be formulated as minimizing the sum of a function and a non-smooth regularization term. Proximal stochastic gradient methods are popular for solving such composite optimization problems. We propose a mini-batch proximal stochastic recursive gradient algorithm SRG-DBB, which incorporates the diagonal Barzilai–Borwein (DBB) stepsize strategy to capture the local geometry of the problem. The linear convergence and complexity of SRG-DBB are analyzed for strongly convex functions. We further establish the linear convergence of SRG-DBB under the non-strong convexity condition. Moreover, it is proved that SRG-DBB converges sublinearly in the convex case. Numerical experiments on standard data sets indicate that the performance of SRG-DBB is better than or comparable to the proximal stochastic recursive gradient algorithm with best-tuned scalar stepsizes or BB stepsizes. Furthermore, SRG-DBB is superior to some advanced mini-batch proximal stochastic gradient methods.
first_indexed 2025-11-14T11:36:30Z
format Journal Article
id curtin-20.500.11937-91425
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T11:36:30Z
publishDate 2022
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-914252023-11-01T04:21:11Z A Mini-Batch Proximal Stochastic Recursive Gradient Algorithm with Diagonal Barzilai–Borwein Stepsize Yu, T.T. Liu, X.W. Dai, Y.H. Sun, Jie Many machine learning problems can be formulated as minimizing the sum of a function and a non-smooth regularization term. Proximal stochastic gradient methods are popular for solving such composite optimization problems. We propose a mini-batch proximal stochastic recursive gradient algorithm SRG-DBB, which incorporates the diagonal Barzilai–Borwein (DBB) stepsize strategy to capture the local geometry of the problem. The linear convergence and complexity of SRG-DBB are analyzed for strongly convex functions. We further establish the linear convergence of SRG-DBB under the non-strong convexity condition. Moreover, it is proved that SRG-DBB converges sublinearly in the convex case. Numerical experiments on standard data sets indicate that the performance of SRG-DBB is better than or comparable to the proximal stochastic recursive gradient algorithm with best-tuned scalar stepsizes or BB stepsizes. Furthermore, SRG-DBB is superior to some advanced mini-batch proximal stochastic gradient methods. 2022 Journal Article http://hdl.handle.net/20.500.11937/91425 10.1007/s40305-022-00436-2 fulltext
spellingShingle Yu, T.T.
Liu, X.W.
Dai, Y.H.
Sun, Jie
A Mini-Batch Proximal Stochastic Recursive Gradient Algorithm with Diagonal Barzilai–Borwein Stepsize
title A Mini-Batch Proximal Stochastic Recursive Gradient Algorithm with Diagonal Barzilai–Borwein Stepsize
title_full A Mini-Batch Proximal Stochastic Recursive Gradient Algorithm with Diagonal Barzilai–Borwein Stepsize
title_fullStr A Mini-Batch Proximal Stochastic Recursive Gradient Algorithm with Diagonal Barzilai–Borwein Stepsize
title_full_unstemmed A Mini-Batch Proximal Stochastic Recursive Gradient Algorithm with Diagonal Barzilai–Borwein Stepsize
title_short A Mini-Batch Proximal Stochastic Recursive Gradient Algorithm with Diagonal Barzilai–Borwein Stepsize
title_sort mini-batch proximal stochastic recursive gradient algorithm with diagonal barzilai–borwein stepsize
url http://hdl.handle.net/20.500.11937/91425