SVD revisited: a new variational principle, compatible feature maps and nonlinear extensions

DOI10.1016/j.acha.2015.09.004zbMath1335.49039OpenAlexW1924769172MaRDI QIDQ262952

Johan A. K. Suykens

Publication date: 4 April 2016

Published in: Applied and Computational Harmonic Analysis (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.acha.2015.09.004

zbMATH Keywords

image processingsingular value decompositionvariational principlenonlinearitykernel methodsLagrange dualityconstrained optimization problemfeature mapsleast squares support vector machines

Mathematics Subject Classification ID

Factorization of matrices (15A23) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08) Duality theory (optimization) (49N15) Discrete approximations in optimal control (49M25) Optimality conditions for problems involving relations other than differential equations (49K21)

Related Items

Deep Restricted Kernel Machines Using Conjugate Feature Duality

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Support-vector networks
• A regularized kernel CCA contrast function for ICA
• Primal and dual model representations in kernel-based learning
• Linear Systems in Self-Adjoint Form
• Learning Theory
• Matrix Analysis
• On the Early History of the Singular Value Decomposition
• Canonical Correlation Analysis: An Overview with Application to Learning Methods
• RELATIONS BETWEEN TWO SETS OF VARIATES