Distribution-free consistency of empirical risk minimization and support vector regression

DOI10.1007/s00498-009-0041-8zbMath1180.93107OpenAlexW2109357635MaRDI QIDQ1047919

Ji-Woong Lee, Pramod P. Khargonekar

Publication date: 8 January 2010

Published in: MCSS. Mathematics of Control, Signals, and Systems (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1007/s00498-009-0041-8

zbMATH Keywords

consistencyweak topologycovering numberPAC learningempirical risk minimizationsupport vector regressiondiscontinuous kernel

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Stochastic learning and adaptive control (93E35)

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Decision theoretic generalizations of the PAC model for neural net and other learning applications
• Aspects of robust linear regression
• Pathologies of some minimum distance estimators
• The ``automatic robustness of minimum distance functionals
• Support vector machines are universally consistent
• Fat-shattering and the learnability of real-valued functions
• Regularization networks and support vector machines
• Some results on Tchebycheffian spline functions and stochastic processes
• On the mathematical foundations of learning
• Training v-Support Vector Regression: Theory and Algorithms
• Covering numbers for real-valued function classes
• Optimal Change‐point Estimation in Inverse Problems
• Scale-sensitive dimensions, uniform convergence, and learnability
• Generalization performance of regularization networks and support vector machines via entropy numbers of compact operators
• Covering numbers for support vector machines
• Structural risk minimization over data-dependent hierarchies
• 10.1162/153244302760200704
• Jump and sharp cusp detection by wavelets
• Choosing multiple parameters for support vector machines