Global convergence of ADMM in nonconvex nonsmooth optimization (Q1736880): Difference between revisions

From MaRDI portal
Importer (talk | contribs)
Changed an Item
Created claim: Wikidata QID (P12): Q129719651, #quickstatements; #temporary_batch_1726362513169
 
(One intermediate revision by one other user not shown)
Property / cites work
 
Property / cites work: Convergence of descent methods for semi-algebraic and tame problems: proximal algorithms, forward-backward splitting, and regularized Gauss-Seidel methods / rank
 
Normal rank
Property / cites work
 
Property / cites work: Optimization with Sparsity-Inducing Penalties / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q3690580 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Practical Augmented Lagrangian Methods for Constrained Optimization / rank
 
Normal rank
Property / cites work
 
Property / cites work: The Łojasiewicz Inequality for Nonsmooth Subanalytic Functions with Applications to Subgradient Dynamical Systems / rank
 
Normal rank
Property / cites work
 
Property / cites work: A Singular Value Thresholding Algorithm for Matrix Completion / rank
 
Normal rank
Property / cites work
 
Property / cites work: Nonconvex Splitting for Regularized Low-Rank + Sparse Decomposition / rank
 
Normal rank
Property / cites work
 
Property / cites work: The direct extension of ADMM for multi-block convex minimization problems is not necessarily convergent / rank
 
Normal rank
Property / cites work
 
Property / cites work: A Globally Convergent Augmented Lagrangian Algorithm for Optimization with General Constraints and Simple Bounds / rank
 
Normal rank
Property / cites work
 
Property / cites work: Complementary pivot theory of mathematical programming / rank
 
Normal rank
Property / cites work
 
Property / cites work: Iteratively reweighted least squares minimization for sparse recovery / rank
 
Normal rank
Property / cites work
 
Property / cites work: Faster Convergence Rates of Relaxed Peaceman-Rachford and ADMM Under Regularity Assumptions / rank
 
Normal rank
Property / cites work
 
Property / cites work: Parallel multi-block ADMM with \(o(1/k)\) convergence / rank
 
Normal rank
Property / cites work
 
Property / cites work: Spectral operators of matrices / rank
 
Normal rank
Property / cites work
 
Property / cites work: A dual algorithm for the solution of nonlinear variational problems via finite element approximation / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q3321366 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q4144736 / rank
 
Normal rank
Property / cites work
 
Property / cites work: On the $O(1/n)$ Convergence Rate of the Douglas–Rachford Alternating Direction Method / rank
 
Normal rank
Property / cites work
 
Property / cites work: Multiplier and gradient methods / rank
 
Normal rank
Property / cites work
 
Property / cites work: Convergence Analysis of Alternating Direction Method of Multipliers for a Family of Nonconvex Problems / rank
 
Normal rank
Property / cites work
 
Property / cites work: ADMM Algorithmic Regularization Paths for Sparse Statistical Machine Learning / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q4938188 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Alternating direction method of multipliers for real and complex polynomial optimization models / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q3231324 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Differences of two semiconvex functions on the real line / rank
 
Normal rank
Property / cites work
 
Property / cites work: A splitting method for orthogonality constrained problems / rank
 
Normal rank
Property / cites work
 
Property / cites work: Global Convergence of Splitting Methods for Nonconvex Composite Optimization / rank
 
Normal rank
Property / cites work
 
Property / cites work: A new relative perturbation theorem for singular subspaces / rank
 
Normal rank
Property / cites work
 
Property / cites work: Parallel Algorithms for Constrained Tensor Factorization via Alternating Direction Method of Multipliers / rank
 
Normal rank
Property / cites work
 
Property / cites work: On semi- and subanalytic geometry / rank
 
Normal rank
Property / cites work
 
Property / cites work: An augmented Lagrangian approach for sparse principal component analysis / rank
 
Normal rank
Property / cites work
 
Property / cites work: On the Convergence of Alternating Direction Lagrangian Methods for Nonconvex Structured Optimization Problems / rank
 
Normal rank
Property / cites work
 
Property / cites work: Semismooth and Semiconvex Functions in Constrained Optimization / rank
 
Normal rank
Property / cites work
 
Property / cites work: The Primal-Dual Hybrid Gradient Method for Semiconvex Splittings / rank
 
Normal rank
Property / cites work
 
Property / cites work: ARock: An Algorithmic Framework for Asynchronous Parallel Coordinate Updates / rank
 
Normal rank
Property / cites work
 
Property / cites work: Prox-regular functions in variational analysis / rank
 
Normal rank
Property / cites work
 
Property / cites work: Variational Analysis / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q3824879 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Augmented Lagrangian alternating direction method for matrix separation based on low-rank factorization / rank
 
Normal rank
Property / cites work
 
Property / cites work: Solving Multiple-Block Separable Convex Minimization Problems Using Two-Block Alternating Direction Method of Multipliers / rank
 
Normal rank
Property / cites work
 
Property / cites work: Linear convergence of adaptively iterative thresholding algorithms for compressed sensing / rank
 
Normal rank
Property / cites work
 
Property / cites work: Characterization of the subdifferential of some matrix norms / rank
 
Normal rank
Property / cites work
 
Property / cites work: Alternating direction methods for classical and ptychographic phase retrieval / rank
 
Normal rank
Property / cites work
 
Property / cites work: A feasible method for optimization with orthogonality constraints / rank
 
Normal rank
Property / cites work
 
Property / cites work: A Block Coordinate Descent Method for Regularized Multiconvex Optimization with Applications to Nonnegative Tensor Factorization and Completion / rank
 
Normal rank
Property / cites work
 
Property / cites work: An alternating direction algorithm for matrix completion with nonnegative factors / rank
 
Normal rank
Property / cites work
 
Property / cites work: Self Equivalence of the Alternating Direction Method of Multipliers / rank
 
Normal rank
Property / cites work
 
Property / cites work: Alternating Direction Method of Multipliers for a Class of Nonconvex and Nonsmooth Problems with Applications to Background/Foreground Extraction / rank
 
Normal rank
Property / cites work
 
Property / cites work: Research on quantum authentication methods for the secure access control among three elements of cloud computing / rank
 
Normal rank
Property / cites work
 
Property / cites work: GAITA: a Gauss-Seidel iterative thresholding algorithm for \(\ell_q\) regularized least squares regression / rank
 
Normal rank
Property / cites work
 
Property / cites work: <formula formulatype="inline"><tex Notation="TeX">$L_{1/2}$</tex> </formula> Regularization: Convergence of Iterative Half Thresholding Algorithm / rank
 
Normal rank
Property / Wikidata QID
 
Property / Wikidata QID: Q129719651 / rank
 
Normal rank

Latest revision as of 03:17, 15 September 2024

scientific article
Language Label Description Also known as
English
Global convergence of ADMM in nonconvex nonsmooth optimization
scientific article

    Statements

    Global convergence of ADMM in nonconvex nonsmooth optimization (English)
    0 references
    0 references
    0 references
    0 references
    26 March 2019
    0 references
    The authors consider the following optimization problem: \[ \left\{ \begin{array}{l}\underset{x_{0},\ldots ,x_{p},y}{\mathrm{min}}\mathrm{\quad }\phi (x_{0},\ldots ,x_{p},y) \\ \text{subject to }\, A_{0}x_{0}+A_{1}x_{1}+\ldots +A_{p}x_{p}+By=0 \end{array}\right. \] where \(\phi \) is a continuous (possibly nonconvex, nonsmooth) function, \(x_{i}\in \mathbb{R}^{n_{i}}\), \(y\in \mathbb{R}^{q}\) and \(A_{i}\), \(B\) are \(m\times n_{i}\) (resp. \(m\times q\)) matrices. The authors propose an ADMM (Alternating Direction Method of Multipliers) type algorithm (multi-block version) that updates sucessively each of the primal variables \(x_{0}, \dots, x_{p}, y\) followed by an update of the dual variable. Under certain assumptions they establish global convergence. This approach applies to problems arising in matrix decomposition, sparse recovery, machine learning and optimization on compact smooth manifolds, covering cases that were not previously covered by any convergence theory. They also discuss relations with the ADL (Augmented Lagrangian Method) and provide an example in which ADMM converges but the latter diverges.
    0 references
    0 references
    nonconvex optimization
    0 references
    alternating direction method of multipliers
    0 references
    augmented Lagrangian method
    0 references
    block coordinate descent
    0 references
    sparse optimization
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references