Linear convergence of accelerated conditional gradient algorithms in spaces of measures
From MaRDI portal
Abstract: A class of generalized conditional gradient algorithms for the solution of optimization problem in spaces of Radon measures is presented. The method iteratively inserts additional Dirac-delta functions and optimizes the corresponding coefficients. Under general assumptions, a sub-linear rate in the objective functional is obtained, which is sharp in most cases. To improve efficiency, one can fully resolve the finite-dimensional subproblems occurring in each iteration of the method. We provide an analysis for the resulting procedure: under a structural assumption on the optimal solution, a linear convergence rate is obtained locally.
Recommendations
- A conditional gradient method with linear rate of convergence for solving convex linear systems
- On the linear convergence rates of exchange and continuous methods for total variation minimization
- On the convergence of the conditional gradient method in distributed optimization problems
- Generalized Nesterov's accelerated proximal gradient algorithms with convergence rate of order o(1/k^2)
- Affine Invariant Convergence Rates of the Conditional Gradient Method
Cites work
- scientific article; zbMATH DE number 3791008 (Why is no real title available?)
- scientific article; zbMATH DE number 3526459 (Why is no real title available?)
- scientific article; zbMATH DE number 204193 (Why is no real title available?)
- scientific article; zbMATH DE number 3345848 (Why is no real title available?)
- A Fast Iterative Shrinkage-Thresholding Algorithm for Linear Inverse Problems
- A Tight Upper Bound on the Rate of Convergence of Frank-Wolfe Algorithm
- A duality-based approach to elliptic control problems in non-reflexive Banach spaces
- A measure space approach to optimal source placement
- A semismooth Newton method with multidimensional filter globalization for l₁-optimization
- A simple proof of Singer’s representation theorem
- A sparse control approach to optimal sensor placement in PDE-constrained parameter estimation problems
- A unifying Radon-Nikodym theorem for vector measures
- An extension of the frank and Wolfe method of feasible directions
- An introduction to -convergence
- Approximation of elliptic control problems in measure spaces with sparse solutions
- Conditional gradient algorithms with open loop step size rules
- Convergence Rates for Conditional Gradient Sequences Generated by Implicit Step Length Rules
- D-optimal designs via a cocktail algorithm
- Design of experiments in nonlinear models. Asymptotic normality, optimality criteria and small-sample properties
- Directional sparsity in optimal control of partial differential equations
- Distributed optimization and statistical learning via the alternating direction method of multipliers
- Elliptic optimal control problems with L^1-control cost and applications for the placement of control devices
- Exact support recovery for sparse spikes deconvolution
- Frequency-sparse optimal quantum control
- Infinite dimensional analysis. A hitchhiker's guide.
- Inverse point source location with the Helmholtz equation on a bounded domain
- Inverse problems in spaces of measures
- Linear convergence of a modified Frank–Wolfe algorithm for computing minimum-volume enclosing ellipsoids
- Measure theory. Vol. I and II
- Measure valued directional sparsity for parabolic optimal control problems
- Model-oriented design of experiments
- Numerical analysis of sparse initial data identification for parabolic problems
- On Choosing and Bounding Probability Metrics
- On linear-quadratic elliptic control problems of semi-infinite type
- Optimal Measurement Methods for Distributed Parameter System Identification
- Optimal control of the undamped linear wave equation with measure valued controls
- Optimum Designs in Regression Problems
- Rates of Convergence for Conditional Gradient Algorithms Near Singular and Nonsingular Extremals
- Second-Order Derivatives of Extremal-Value Functions and Optimality Conditions for Semi-Infinite Programs
- Semi-Infinite Programming: Theory, Methods, and Applications
- Semismooth Newton Methods for Operator Equations in Function Spaces
- Some algorithmic aspects of the theory of optimal designs
- Some iterative procedures for generating nonsingular optimal designs
- Sparse inverse problems over measures: equivalence of the conditional gradient and exchange methods
- Sparsity of solutions for variational inverse problems with finite-dimensional data
- Spike controls for elliptic and parabolic PDEs
- Spike detection from inaccurate samplings
- The Primal-Dual Active Set Strategy as a Semismooth Newton Method
- The Sequential Generation of D-Optimum Experimental Designs
- The alternating descent conditional gradient method for sparse inverse problems
- The length of the primal-dual path in Moreau-Yosida-based path-following methods for state constrained optimal control
- The sliding Frank-Wolfe algorithm and its application to super-resolution microscopy
- Towards a Mathematical Theory of Super‐resolution
Cited in
(18)- Asymptotic linear convergence of fully-corrective generalized conditional gradient methods
- A generalized conditional gradient method for dynamic inverse problems with optimal transport regularization
- Proximal methods for point source localisation
- A conditional gradient method with linear rate of convergence for solving convex linear systems
- Sparse inverse problems over measures: equivalence of the conditional gradient and exchange methods
- Nonconvex regularization for sparse neural networks
- Grid is good. Adaptive refinement algorithms for off-the-grid total variation minimization
- A sparse optimization approach to infinite infimal convolution regularization
- Convergence analysis of the discretization of continuous-domain inverse problems
- No-gap second-order conditions for minimization problems in spaces of measures
- Linearly convergent away-step conditional gradient for non-strongly convex functions
- On non-autonomous parabolic equations with measure-valued right hand sides and applications to optimal control
- On the linear convergence rates of exchange and continuous methods for total variation minimization
- Convergence rates of gradient methods for convex optimization in the space of measures
- Sparse optimization on measures with over-parameterized gradient descent
- scientific article; zbMATH DE number 4057300 (Why is no real title available?)
- scientific article; zbMATH DE number 7064051 (Why is no real title available?)
- A Linearly Convergent Variant of the Conditional Gradient Algorithm under Strong Convexity, with Applications to Online and Stochastic Optimization
This page was built for publication: Linear convergence of accelerated conditional gradient algorithms in spaces of measures
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4999535)
