The Douglas-Rachford algorithm converges only weakly

DOI10.1137/19M1308451MaRDI QIDQ5107064zbMATH OpenOpenAlexFDO

Authors Minh N. Bui, Patrick L. Combettes

Publication date 22 April 2020

Published in SIAM Journal on Control and Optimization (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1912.09564

zbMATH Keywords

Douglas-Rachford algorithmmethod of partial inversesmonotone operatoroperator splittingstrong convergenceweak convergence

Mathematics Subject Classification ID

Monotone operators and generalizations (47H05) Contraction-type mappings, nonexpansive mappings, (A)-proper mappings, etc. (47H09) Iterative procedures involving nonlinear operators (47J25)

Abstract: We show that the weak convergence of the Douglas--Rachford algorithm for finding a zero of the sum of two maximally monotone operators cannot be improved to strong convergence. Likewise, we show that strong convergence can fail for the method of partial inverses.

Recommendations

• On weak convergence of the Douglas-Rachford method
• The Douglas-Rachford algorithm in the affine-convex case
• A simplified proof of weak convergence in Douglas-Rachford method
• On the local convergence of the Douglas-Rachford algorithm
• On the order of the operators in the Douglas-Rachford algorithm

Cites work

• An alternating projection that does not converge in norm
• Convex analysis and monotone operator theory in Hilbert spaces
• Iterative construction of the resolvent of a sum of maximal monotone operators
• On the Numerical Solution of Heat Conduction Problems in Two and Three Space Variables
• On weak convergence of the Douglas-Rachford method
• Partial inverse of a monotone operator
• Splitting Algorithms for the Sum of Two Nonlinear Operators

Cited in

(9)

• Demiclosedness principles for generalized nonexpansive mappings
• A simplified proof of weak convergence in Douglas-Rachford method
• On weak convergence of the Douglas-Rachford method
• Proximal Splitting Algorithms for Convex Optimization: A Tour of Recent Advances, with New Twists
• Randomized Douglas–Rachford Methods for Linear Systems: Improved Accuracy and Efficiency
• Resolvent and proximal compositions
• The geometry of monotone operator splitting methods
• HPPP: Halpern-type preconditioned proximal point algorithms and applications to image restoration
• Frugal and decentralised resolvent splittings defined by nonexpansive operators

This page was built for publication: The Douglas-Rachford algorithm converges only weakly

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5107064)