Orthogonal invariance and identifiability

DOI10.1137/130916710MaRDI QIDQ2923360zbMATH OpenOpenAlexFDO

Authors A. Daniilidis, D. Drusvyatskiy, A. S. Lewis

Publication date 15 October 2014

Published in SIAM Journal on Matrix Analysis and Applications (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1304.1198

zbMATH Keywords

dualityeigenvaluesidentifiable setoptimization algorithmpartial smoothnesspolyhedrasemidefinite programmingsymmetric matrix

Mathematics Subject Classification ID

Eigenvalues, singular values, and eigenvectors (15A18) Factorization of matrices (15A23) Semidefinite programming (90C22) Convex programming (90C25)

Abstract: Orthogonally invariant functions of symmetric matrices often inherit properties from their diagonal restrictions: von Neumann's theorem on matrix norms is an early example. We discuss the example of "identifiability", a common property of nonsmooth functions associated with the existence of a smooth manifold of approximate critical points. Identifiability (or its synonym, "partial smoothness") is the key idea underlying active set methods in optimization. Polyhedral functions, in particular, are always partly smooth, and hence so are many standard examples from eigenvalue optimization.

Recommendations

• Group Invariance and Convex Matrix Analysis
• Optimality, identifiability, and sensitivity
• Variational analysis of spectral functions simplified
• Semidefinite optimization
• On differentiability of a class of orthogonally invariant functions on several operator variables

Cited in

(25)

• The Euclidean distance degree of orthogonally invariant matrix varieties
• On the connectedness of spectral sets and irreducibility of spectral cones in Euclidean Jordan algebras
• BLIM's identifiability and parameter invariance under backward and forward transformations
• Sharp oracle inequalities for low-complexity priors
• Local convergence properties of Douglas-Rachford and alternating direction method of multipliers
• The degrees of freedom of partly smooth regularizers
• Spectral (isotropic) manifolds and their dimension
• Low complexity regularization of linear inverse problems
• Activity identification and local linear convergence of forward-backward-type methods
• Counting real critical points of the distance to orthogonally invariant matrix sets
• Activity identification and local linear convergence of Douglas-Rachford/ADMM under partial smoothness
• scientific article; zbMATH DE number 3949535 (Why is no real title available?)
• Sensitivity analysis for mirror-stratifiable convex functions
• Variational analysis of spectral functions simplified
• Optimality, identifiability, and sensitivity
• A unified approach to spectral and isotropic functions
• Extreme point inequalities and geometry of the rank sparsity ball
• Derivatives of compound matrix valued functions
• Spectral sets and functions on Euclidean Jordan algebras
• Spectral operators of matrices: semismoothness and characterizations of the generalized Jacobian
• Approximating functions on stratified sets
• A complexity analysis framework for active manifold identification with applications to L₀ and L_p regularization models
• On the geometry and refined rate of primal-dual hybrid gradient for linear programming
• Active manifolds, stratifications, and convergence to local minima in nonsmooth optimization
• The geometry and well-posedness of sparse regularized linear regression

This page was built for publication: Orthogonal invariance and identifiability

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2923360)