The computational intractability of training sigmoidal neural networks

DOI10.1109/18.567673zbMath0874.68255OpenAlexW2108030384MaRDI QIDQ4336293

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

zbMATH Keywords

neural networkNP-completenessLipschitzian sigmoidal internal activation functions

Mathematics Subject Classification ID

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

Related Items

Ridgelets: estimating with ridge functions, CONVERGENCE OF A LEAST‐SQUARES MONTE CARLO ALGORITHM FOR AMERICAN OPTION PRICING WITH DEPENDENT SAMPLE DATA, Training a Single Sigmoidal Neuron Is Hard, Some problems in the theory of ridge functions, Local greedy approximation for nonlinear regression and neural network training., On approximate learning by multi-layered feedforward circuits, Hardness results for neural network approximation problems