Stochastic subspace correction in Hilbert space

DOI10.1007/S00365-018-9447-1MaRDI QIDQ1615986zbMATH OpenOpenAlexFDO

Authors Peter Oswald, Michael Griebel

Publication date 31 October 2018

Published in Constructive Approximation (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1712.02111

zbMATH Keywords

block coordinate descentconvergence ratesgreedyinfinite space splittingmultiplicative Schwarzonline learningrandomizedsubspace correction

Mathematics Subject Classification ID

Decomposition methods (49M27) Iterative numerical methods for linear systems (65F10) Numerical solution of discretized equations for boundary value problems involving PDEs (65N22)

Abstract: We consider an incremental approximation method for solving variational problems in infinite-dimensional Hilbert spaces, where in each step a randomly and independently selected subproblem from an infinite collection of subproblems is solved. we show that convergence rates for the expectation of the squared error can be guaranteed under weaker conditions than previously established in [Constr. Approx. 44:1 (2016), 121-139]. A connection to the theory of learning algorithms in reproducing kernel Hilbert spaces is revealed.

Recommendations

• Stochastic subspace correction methods and fault tolerance
• Sub-linear convergence of a stochastic proximal iteration method in Hilbert space
• Finite dimensional approximation and Newton-based algorithm for stochastic approximation in Hilbert space
• An invariance principle for a finite dimensional stochastic approximation method in a Hilbert space
• Schwarz iterative methods: infinite space splittings

Cites work

• Approximation and learning by greedy algorithms
• Greedy Algorithms for Optimal Measurements Selection in State Estimation Using Reduced Models
• Greedy and randomized versions of the multiplicative Schwarz method
• Greedy approximation
• Iterative Methods by Space Decomposition and Subspace Correction
• Learning Theory
• Learning theory of randomized Kaczmarz algorithm
• Nonparametric stochastic approximation with large step-sizes
• On the abstract theory of additive and multiplicative Schwarz algorithms
• On the small balls problem for equivalent Gaussian measures
• Online Learning as Stochastic Approximation of Regularization Paths: Optimality and Almost-Sure Convergence
• Optimization in high dimensions via accelerated, parallel, and proximal coordinate descent
• Scattered Data Approximation
• Schwarz iterative methods: infinite space splittings
• Stochastic Equations in Infinite Dimensions

Cited in

(4)

• Rates of convergence of randomized Kaczmarz algorithms in Hilbert spaces
• Greedy Algorithms for Optimal Measurements Selection in State Estimation Using Reduced Models
• Stochastic subspace correction methods and fault tolerance
• Convergence analysis of online algorithms for vector-valued kernel regression

This page was built for publication: Stochastic subspace correction in Hilbert space

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1615986)