Convergence Results for Projected Line-Search Methods on Varieties of Low-Rank Matrices Via Łojasiewicz Inequality

DOI10.1137/140957822zbMath1355.65079arXiv1402.5284OpenAlexW1993468393MaRDI QIDQ2954392

Reinhold Schneider, André Uschmajew

Publication date: 13 January 2017

Published in: SIAM Journal on Optimization (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1402.5284

zbMATH Keywords

convergence analysissteepest descenttangent conesRiemannian optimizationline search methodslow-rank matricesŁojasiewicz gradient inequality

Mathematics Subject Classification ID

Numerical mathematical programming methods (65K05) Nonlinear programming (90C30) Programming in abstract spaces (90C48)

Related Items (54)

Convergence rate analysis for the higher order power method in best rank one approximations of tensors ⋮ Analysis of Asymptotic Escape of Strict Saddle Sets in Manifold Optimization ⋮ Adaptive stochastic Galerkin FEM with hierarchical tensor representations ⋮ The Optimization Landscape for Fitting a Rank-2 Tensor with a Rank-1 Tensor ⋮ Tangent cones to tensor train varieties ⋮ Tensor Completion in Hierarchical Tensor Representations ⋮ Tensor Canonical Correlation Analysis With Convergence and Statistical Guarantees ⋮ Error bound conditions and convergence of optimization methods on smooth and proximally smooth manifolds ⋮ Stability of minimization problems and the error bound condition ⋮ Alternating Least Squares as Moving Subspace Correction ⋮ Convergence Analysis on SS-HOPM for BEC-Like Nonlinear Eigenvalue Problems ⋮ On the continuity of the tangent cone to the determinantal variety ⋮ Sketching for a low-rank nonnegative matrix approximation: numerical study ⋮ Finding stationary points on bounded-rank matrices: a geometric hurdle and a smooth remedy ⋮ Jacobi-type algorithms for homogeneous polynomial optimization on Stiefel manifolds with applications to tensor approximations ⋮ Low-rank tensor methods for partial differential equations ⋮ An Apocalypse-Free First-Order Low-Rank Optimization Algorithm with at Most One Rank Reduction Attempt per Iteration ⋮ Unnamed Item ⋮ A nonconvex approach to low-rank matrix completion using convex optimization ⋮ Convergence of Gradient-Based Block Coordinate Descent Algorithms for Nonorthogonal Joint Approximate Diagonalization of Matrices ⋮ Normal Cones Intersection Rule and Optimality Analysis for Low-Rank Matrix Optimization with Affine Manifolds ⋮ Riemannian thresholding methods for row-sparse and low-rank matrix recovery ⋮ Linear Convergence of a Proximal Alternating Minimization Method with Extrapolation for \(\boldsymbol{\ell_1}\) -Norm Principal Component Analysis ⋮ An approximate projection onto the tangent cone to the variety of third-order tensors of bounded tensor-train rank ⋮ Optimality conditions for Tucker low-rank tensor optimization ⋮ A global exact penalty for rank-constrained optimization problem and applications ⋮ Gradient projection method on matrix manifolds ⋮ On the gradient projection method for weakly convex functions on a proximally smooth set ⋮ Approximate Matrix and Tensor Diagonalization by Unitary Transformations: Convergence of Jacobi-Type Algorithms ⋮ Descent line search scheme using Geršgorin circle theorem ⋮ Sampling-free Bayesian inversion with adaptive hierarchical tensor representations ⋮ Mini-workshop: Computational optimization on manifolds. Abstracts from the mini-workshop held November 15--21, 2020 (online meeting) ⋮ Tensor networks and hierarchical tensors for the solution of high-dimensional partial differential equations ⋮ RSG: Beating Subgradient Method without Smoothness and Strong Convexity ⋮ Computing Riemannian center of mass on Hadamard manifolds ⋮ The point-wise convergence of shifted symmetric higher order power method ⋮ Growth conditions on a function and the error bound condition ⋮ Characterization of solution sets of convex optimization problems in Riemannian manifolds ⋮ The gradient projection algorithm for smooth sets and functions in nonconvex case ⋮ Matrix optimization over low-rank spectral sets: stationary points and local and global minimizers ⋮ The gradient projection method with Armijo's step size on manifolds ⋮ Correction to: ``An approximate augmented Lagrangian method for nonnegative low-rank matrix approximation ⋮ Optimality conditions for rank-constrained matrix optimization ⋮ Riemannian Optimization for High-Dimensional Tensor Completion ⋮ A Riemannian rank-adaptive method for low-rank matrix completion ⋮ A Gradient Sampling Method on Algebraic Varieties and Application to Nonsmooth Low-Rank Optimization ⋮ Quadratic optimization with orthogonality constraint: explicit Łojasiewicz exponent and linear convergence of retraction-based line-search and stochastic variance-reduced gradient methods ⋮ A Riemannian gradient ascent algorithm with applications to orthogonal approximation problems of symmetric tensors ⋮ On the geometric analysis of a quartic-quadratic optimization problem under a spherical constraint ⋮ Globally Convergent Jacobi-Type Algorithms for Simultaneous Orthogonal Symmetric Tensor Diagonalization ⋮ Tensor Spaces and Hierarchical Tensor Representations ⋮ New Riemannian Preconditioned Algorithms for Tensor Completion via Polyadic Decomposition ⋮ Optimization on the hierarchical Tucker manifold - applications to tensor completion ⋮ Structured low-rank approximation: optimization on matrix manifold approach

Uses Software

• SDPLR

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Tensor Decompositions and Applications
• Tensor-Train Decomposition
• The geometry of algorithms using hierarchical tensors
• Low-rank tensor completion by Riemannian optimization
• Greedy algorithms for high-dimensional eigenvalue problems
• The Łojasiewicz gradient inequality in the infinite-dimensional Hilbert space framework
• On the convergence of the proximal algorithm for nonsmooth functions involving analytic features
• Dynamical low-rank approximation: Applications and numerical experiments
• Numerical computation of an analytic singular value decomposition of a matrix valued function
• On gradients of functions definable in o-minimal structures
• Critical points of matrix least squares distance functions
• A nonlinear programming algorithm for solving semidefinite programs via low-rank factorization
• Iterative methods for low rank approximation of graph similarity matrices
• Convergence of descent methods for semi-algebraic and tame problems: proximal algorithms, forward-backward splitting, and regularized Gauss-Seidel methods
• Prox-regularity of rank constraint sets and implications for algorithms
• Convergence of non-smooth descent methods using the Kurdyka-Łojasiewicz inequality
• Fixed-rank matrix factorizations and Riemannian low-rank optimization
• A new scheme for the tensor representation
• Convergence to equilibrium for discretizations of gradient-like flows on Riemannian manifolds
• A projector-splitting integrator for dynamical low-rank approximation
• Convergence of gradient-based algorithms for the Hartree-Fock equations
• Low-Rank Matrix Completion by Riemannian Optimization
• Dynamical Approximation by Hierarchical Tucker and Tensor-Train Tensors
• A Block Coordinate Descent Method for Regularized Multiconvex Optimization with Applications to Nonnegative Tensor Factorization and Completion
• Projection-like Retractions on Matrix Manifolds
• On Convergence of the Maximum Block Improvement Method
• Dynamical Tensor Approximation
• A Riemannian Optimization Approach for Computing Low-Rank Solutions of Lyapunov Equations
• Low-Rank Optimization on the Cone of Positive Semidefinite Matrices
• Proximal Alternating Minimization and Projection Methods for Nonconvex Problems: An Approach Based on the Kurdyka-Łojasiewicz Inequality
• Clarke Subgradients of Stratifiable Functions
• Characterizations of Łojasiewicz inequalities: Subgradient flows, talweg, convexity
• Variational Analysis
• Limits of tangent spaces to real surfaces
• Low-Rank Optimization with Trace Norm Penalty
• The Łojasiewicz Inequality for Nonsmooth Subanalytic Functions with Applications to Subgradient Dynamical Systems
• Pointwise convergence of gradient‐like systems
• Dynamical Low‐Rank Approximation
• Convergence of the Iterates of Descent Methods for Analytic Cost Functions
• Generalized Kuhn–Tucker Conditions for Mathematical Programming Problems in a Banach Space

This page was built for publication: Convergence Results for Projected Line-Search Methods on Varieties of Low-Rank Matrices Via Łojasiewicz Inequality