Training a Single Sigmoidal Neuron Is Hard

DOI10.1162/089976602760408035zbMATH Open1060.68099DBLPjournals/neco/Sima02OpenAlexW2171931454WikidataQ52112387 ScholiaQ52112387MaRDI QIDQ4409384FDOQ4409384

Authors: Jiří Šíma 

Publication date: 14 October 2003

Published in: Neural Computation (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1162/089976602760408035

Recommendations

• scientific article; zbMATH DE number 2080651
• The computational intractability of training sigmoidal neural networks
• scientific article; zbMATH DE number 2080455
• Computational limitations on training sigmoid neural networks
• Hardness results for neural network approximation problems

zbMATH Keywords

feedforward neural networksconstructive learninghardness results

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Cites Work

• Learning representations by back-propagating errors
• Learnability and the Vapnik-Chervonenkis dimension
• On the complexity of polyhedral separability
• Feedforward nets for interpolation and classification
• Computational limitations on learning from examples
• The densest hemisphere problem
• The hardness of approximate optima in lattices, codes, and systems of linear equations
• Robust trainability of single neurons
• Computational limitations on training sigmoid neural networks
• On the complexity of loading shallow neural networks
• On the geometric separability of Boolean functions
• The computational intractability of training sigmoidal neural networks

Cited In (15)

• Title not available (Why is that?)
• Computational limitations on training sigmoid neural networks
• Recovery of regular ridge functions on the ball
• Title not available (Why is that?)
• Black holes and the loss landscape in machine learning
• Loading Deep Networks Is Hard: The Pyramidal Case
• Global optimization issues in deep network regression: an overview
• On the omnipresence of spurious local minima in certain neural network training problems
• Complexity of network training for classes of Neural Networks
• On the Nonlearnability of a Single Spiking Neuron
• fPINNs: Fractional Physics-Informed Neural Networks
• Complexity of training ReLU neural network
• Title not available (Why is that?)
• First Steps Towards a Runtime Analysis of Neuroevolution
• Long-Range Out-of-Sample Properties of Autoregressive Neural Networks

This page was built for publication: Training a Single Sigmoidal Neuron Is Hard

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4409384)