The importance of convexity in learning with squared loss

DOI10.1109/18.705577MaRDI QIDQ4701173zbMATH OpenOpenAlexFDO

Authors Wee Sun Lee, Peter L. Bartlett, Robert C. Williamson

Publication date 21 November 1999

Published in IEEE Transactions on Information Theory (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1109/18.705577

zbMATH Keywords

agnostic learning

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Recommendations

• Correction to "The Importance of Convexity in Learning With Squared Loss" [Sep 98 1974-1980]
• Learning with convex loss and indefinite kernels
• scientific article; zbMATH DE number 7306907
• Convexity Properties of the Erlang Loss Formula
• Iterative regularization for learning with convex loss functions
• scientific article; zbMATH DE number 7626778
• Convexity, Classification, and Risk Bounds
• From inexact optimization to learning via gradient concentration
• Quasi-convex loss functions, unimodal distributions, and the convolution theorem

Cited in

(28)

• Boosting the margin: a new explanation for the effectiveness of voting methods
• Efficient algorithms for learning functions with bounded variation
• General oracle inequalities for model selection
• On universal estimators in learning theory
• Fast learning from \(\alpha\)-mixing observations
• Local Rademacher complexities
• On the mathematical foundations of learning
• The sample complexity of learning linear predictors with the squared loss
• On martingale extensions of Vapnik-Chervonenkis theory with applications to online learning
• Tikhonov, Ivanov and Morozov regularization for support vector machine learning
• scientific article; zbMATH DE number 2089350 (Why is no real title available?)
• On the optimality of sample-based estimates of the expectation of the empirical minimizer
• Large-scale machine learning with stochastic gradient descent
• Theory of Classification: a Survey of Some Recent Advances
• Convexity Properties of the Erlang Loss Formula
• Correction to "The Importance of Convexity in Learning With Squared Loss" [Sep 98 1974-1980]
• General nonexact oracle inequalities for classes with a subexponential envelope
• Optimization methods for large-scale machine learning
• scientific article; zbMATH DE number 7306907 (Why is no real title available?)
• An unrestricted learning procedure
• Learning Bounds for Kernel Regression Using Effective Data Dimensionality
• Nonexact oracle inequalities, r-learnability, and fast rates
• Sharper lower bounds on the performance of the empirical risk minimization algorithm
• A quadratic sample complexity reduction for agnostic learning via quantum algorithms
• Generalization bounds: perspectives from information theory and PAC-Bayes
• Aggregation via empirical risk minimization
• Regularization in kernel learning
• Obtaining fast error rates in nonconvex situations

This page was built for publication: The importance of convexity in learning with squared loss

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4701173)