A homotopy method for training neural networks

DOI10.1016/S0165-1684(97)00201-6MaRDI QIDQ1129237zbMATH OpenOpenAlexFDO

Authors Markus Lendl, Rolf Unbehauen, Fa-Long Luo

Publication date 13 August 1998

Published in Signal Processing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/s0165-1684(97)00201-6

zbMATH Keywords

homotopyGauss-Newton methodmultilayer perceptronsecond-order learning

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Signal theory (characterization, reconstruction, filtering, etc.) (94A12)

Recommendations

• A homotopy training algorithm for fully connected neural networks
• Artificial Intelligence and Soft Computing - ICAISC 2004
• scientific article; zbMATH DE number 1054676
• Training feedforward networks using simultaneous perturbation with dynamic tunneling
• Ill-Conditioning in Neural Network Training Problems

Cited in

(5)

• Training feedforward networks using simultaneous perturbation with dynamic tunneling
• A homotopy training algorithm for fully connected neural networks
• Homomorphisms preserving neural ideals
• A Homomorphic Neural Network for Modeling and Prediction
• Homogeneous alternating projection neural networks

This page was built for publication: A homotopy method for training neural networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1129237)