Trainability and Accuracy of Artificial Neural Networks: An Interacting Particle System Approach
- 21 July 2022
- journal article
- research article
- Published by Wiley in Communications on Pure and Applied Mathematics
- Vol. 75 (9), 1889-1935
- https://doi.org/10.1002/cpa.22074
Abstract
No abstract availableKeywords
This publication has 28 references indexed in Scilit:
- Random Matrices and Complexity of Spin GlassesCommunications on Pure and Applied Mathematics, 2012
- Generalized Neural-Network Representation of High-Dimensional Potential-Energy SurfacesPhysical Review Letters, 2007
- On‐line learning for very large data setsApplied Stochastic Models in Business and Industry, 2005
- Langevin equation for the density of a system of interacting Langevin processesJournal of Physics A: General Physics, 1996
- Universal approximation bounds for superpositions of a sigmoidal functionIEEE Transactions on Information Theory, 1993
- Universal Approximation Using Radial-Basis-Function NetworksNeural Computation, 1991
- Approximation by superpositions of a sigmoidal functionMathematics of Control, Signals, and Systems, 1989
- On the McKean‐Vlasov Limit for Interacting DiffusionsMathematische Nachrichten, 1988
- Large deviations from the mckean-vlasov limit for weakly interacting diffusionsStochastics, 1987
- A CLASS OF MARKOV PROCESSES ASSOCIATED WITH NONLINEAR PARABOLIC EQUATIONSProceedings of the National Academy of Sciences of the United States of America, 1966