Learning a sparse representation of Barron functions with the inverse scale space flow

DOI10.4208/JML.240123MaRDI QIDQ6955699zbMATH OpenFDO

Authors Tjeerd Jan Heeringa, Tim Roith, Christoph Brune, Martin Burger

Publication date 22 July 2025

Published in Journal of Machine Learning (Search for Journal in Brave)

Full work available at URL https://doi.org/10.4208/JML.240123

zbMATH Keywords

Barron spaceBregman iterationsinverse scale spaceoptimizationsparse neural networks

Mathematics Subject Classification ID

Numerical analysis in abstract spaces (65Jxx) Numerical methods for mathematical programming, optimization and variational techniques (65Kxx) Mathematical programming (90Cxx)

Cites work

• A handbook of -convergence
• A Nonlinear Inverse Scale Space Method for a Convex Multiplicative Noise Model
• An adaptive inverse scale space method for compressed sensing
• An Iterative Regularization Method for Total Variation-Based Image Restoration
• Analysis and generalizations of the linearized Bregman method
• Banach space representer theorems for neural networks and ridge splines
• Bregman Iterative Algorithms for \ell₁-Minimization with Applications to Compressed Sensing
• Choose your path wisely: gradient descent in a Bregman distance framework
• Convergence of the linearized Bregman iteration for \(\ell _1\)-norm minimization
• Convergence rates of convex variational regularization
• Embeddings between Barron spaces with higher-order activation functions
• Error estimation for Bregman iterations and inverse scale space methods in image restoration
• scientific article; zbMATH DE number 3528986 (Why is no real title available?)
• scientific article; zbMATH DE number 845714 (Why is no real title available?)
• scientific article; zbMATH DE number 3398324 (Why is no real title available?)
• Inverse problems in spaces of measures
• Iterative total variation schemes for nonlinear inverse problems
• Linearized Bregman iterations for compressed sensing
• Mirror descent and nonlinear projected subgradient methods for convex optimization.
• Modern regularization methods for inverse problems
• Multiscale methods for polyhedral regularizations
• Nonlinear inverse scale space methods
• On the Banach spaces associated with multi-layer ReLU networks: function representation, approximation theory and gradient descent dynamics
• On the measure of Voronoi cells
• Primal and dual Bregman methods with application to optical nanoscopy
• Representation formulas and pointwise properties for Barron functions
• The Barron space and the flow-induced function spaces for neural network models
• Understanding neural networks with reproducing kernel Banach spaces
• What Kinds of Functions Do Deep Neural Networks Learn? Insights from Variational Spline Theory

This page was built for publication: Learning a sparse representation of Barron functions with the inverse scale space flow

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6955699)