A class of adaptive learning rate algorithms derived by one-dimensional subminimization methods

MaRDI QIDQ2746460zbMATH OpenFDO

Authors George D. Magoulas, Michael N. Vrahatis

Publication date 21 February 2002

Published in Neural, Parallel \& Scientific Computations (Search for Journal in Brave)

zbMATH Keywords

adaptive learning rate algorithmsbackpropagation networks

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Parallel algorithms in computer science (68W10)

Recommendations

• Acceleration of the back propagation through dynamic adaptation of the learning rate
• Learning rate adaptation in stochastic gradient descent.
• A class of asymptotically stable algorithms for learning-rate adaptation
• scientific article; zbMATH DE number 1424367
• ADAPTIVE ALGORITHMS FOR NEURAL NETWORK SUPERVISED LEARNING: A DETERMINISTIC OPTIMIZATION APPROACH

Cited in

(8)

• A class of asymptotically stable algorithms for learning-rate adaptation
• Globally convergent modification of the quickprop method
• Improved sign-based learning algorithm derived by the composite nonlinear Jacobi process
• New classes of learning automata based schemes for adaptation of backpropagation algorithm parameters
• scientific article; zbMATH DE number 1304442 (Why is no real title available?)
• scientific article; zbMATH DE number 1928683 (Why is no real title available?)
• scientific article; zbMATH DE number 7583836 (Why is no real title available?)
• An adaptive adjustment method of learning rate in logistic regression

This page was built for publication: A class of adaptive learning rate algorithms derived by one-dimensional subminimization methods

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2746460)