Stochastic incremental mirror descent algorithms with Nesterov smoothing

DOI10.1007/S11075-023-01574-1MaRDI QIDQ6145577zbMATH OpenOpenAlexFDO

Authors Sandy Bitterlich, Sorin-Mihai Grad

Publication date 9 January 2024

Published in Numerical Algorithms (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/s11075-023-01574-1

zbMATH Keywords

mirror descentNesterov smoothingstochastic

Mathematics Subject Classification ID

Numerical solutions to stochastic differential and integral equations (65C30) Numerical aspects of computer graphics, image analysis, and computational geometry (65D18)

Review text

This paper presents two incremental stochastic mirror descent algorithms for minimizing sums of finitely many non-smooth convex functions over convex sets. The Nesterov smoothing technique is used for the gradients of the smooth summands of the objective function instead of their subgradients. The functions are not required to be Lipschitz continuous or differentiable. Three algorithms are considered including an optimization problem in classifying images using support vector machines, a problem in tomography and a continuous location problem.

Recommendations

• Mirror descent and convex optimization problems with non-smooth inequality constraints
• An incremental mirror descent subgradient algorithm with random sweeping and proximal step
• Stochastic block mirror descent methods for nonsmooth and stochastic optimization
• An inexact gradient mirror descent algorithm for non-smooth convex optimization
• Gradient-free proximal methods with inexact oracle for convex stochastic nonsmooth optimization problems on the simplex

Cites work

• A proximal method for solving nonlinear minmax location problems with perturbed minimal time functions via conjugate duality
• A version of the mirror descent method to solve variational inequalities
• Accelerated randomized mirror descent algorithms for composite non-strongly convex optimization
• Adaptive smoothing algorithms for nonsmooth composite convex minimization
• Analogues of switching subgradient schemes for relatively Lipschitz-continuous convex programming problems
• Convex Analysis
• Duality between subgradient and conditional gradient methods
• Ergodic mirror descent
• Fastest rates for stochastic mirror descent methods
• First-order methods in optimization
• scientific article; zbMATH DE number 3695967 (Why is no real title available?)
• scientific article; zbMATH DE number 3790208 (Why is no real title available?)
• Inexact stochastic mirror descent for two-stage nonlinear stochastic programs
• Lectures on convex optimization
• Linear coupling: an ultimate unification of gradient and mirror descent
• MAGMA: multilevel accelerated gradient mirror descent algorithm for large-scale convex composite minimization
• Mirror descent and nonlinear projected subgradient methods for convex optimization.
• On stochastic subgradient mirror-descent algorithm with weighted averaging
• On the convergence of mirror descent beyond stochastic convex programming
• Optimal subgradient methods: computational properties for large-scale linear inverse problems
• Primal-dual subgradient methods for convex problems
• Smooth minimization of non-smooth functions
• Smoothing and first order methods: a unified framework
• Smoothing techniques and difference of convex functions algorithms for image reconstructions
• Stochastic mirror descent dynamics and their convergence in monotone variational inequalities
• The ordered subsets mirror descent optimization method with applications to tomography
• Unifying mirror descent and dual averaging
• Variable smoothing for convex optimization problems using stochastic gradients

Cited in

(1)

• An inexact gradient mirror descent algorithm for non-smooth convex optimization

This page was built for publication: Stochastic incremental mirror descent algorithms with Nesterov smoothing

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6145577)