The computational intractability of training sigmoidal neural networks

DOI10.1109/18.567673zbMATH Open0874.68255OpenAlexW2108030384MaRDI QIDQ4336293FDOQ4336293

Authors: Lee Kenneth Jones 

Publication date: 10 November 1997

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

Full work available at URL: https://doi.org/10.1109/18.567673

Recommendations

• Training a Single Sigmoidal Neuron Is Hard
• On the infeasibility of training neural networks with small mean-squared error
• Computational limitations on training sigmoid neural networks
• scientific article; zbMATH DE number 2080651

zbMATH Keywords

neural networkNP-completenessLipschitzian sigmoidal internal activation functions

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Circuits, networks (94C99)

Cited In (19)

• Ill-Conditioning in Neural Network Training Problems
• A mathematical solution to a network construction problem.
• Computational limitations on training sigmoid neural networks
• Title not available (Why is that?)
• Title not available (Why is that?)
• An unfeasibility view of neural network learning
• On approximate learning by multi-layered feedforward circuits
• Algorithmic Learning Theory
• Title not available (Why is that?)
• Loading Deep Networks Is Hard: The Pyramidal Case
• Local greedy approximation for nonlinear regression and neural network training.
• Training a Single Sigmoidal Neuron Is Hard
• Title not available (Why is that?)
• Some problems in the theory of ridge functions
• Ridgelets: estimating with ridge functions
• Hardness results for neural network approximation problems
• On the complexity of loading shallow neural networks
• Title not available (Why is that?)
• CONVERGENCE OF A LEAST‐SQUARES MONTE CARLO ALGORITHM FOR AMERICAN OPTION PRICING WITH DEPENDENT SAMPLE DATA

This page was built for publication: The computational intractability of training sigmoidal neural networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4336293)