Memory capacity of two layer neural networks with smooth activations

DOI10.1137/23M1599355MaRDI QIDQ6583717zbMATH OpenFDO

Authors Liam Madden, Christos Thrampoulidis

Publication date 6 August 2024

Published in SIAM Journal on Mathematics of Data Science (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1137/23M1599355

zbMATH Keywords

Hadamard powersKhatri-Rao productsKruskal rankmemory capacityneural networks

Mathematics Subject Classification ID

Computer science (68-XX)

Cites work

• Enumeration of Seven-Argument Threshold Functions
• Hadamard powers and kernel perceptrons
• scientific article; zbMATH DE number 1235881 (Why is no real title available?)
• scientific article; zbMATH DE number 3998214 (Why is no real title available?)
• scientific article; zbMATH DE number 3227305 (Why is no real title available?)
• Identifiability of parameters in latent structure models with many observed variables
• Introduction to Smooth Manifolds
• Linear algebra done right
• On the capabilities of multilayer perceptrons
• Perturbed Identity Matrices Have High Rank: Proof and Applications
• Reconciling modern machine-learning practice and the classical bias-variance trade-off
• The interpolation phase transition in neural networks: memorization and generalization under lazy training
• Three-way arrays: rank and uniqueness of trilinear decompositions, with application to arithmetic complexity and statistics
• User-friendly tail bounds for sums of random matrices

Cited in

(1)

• Representation of the solution of fractional Schrödinger equations with shallow neural network

This page was built for publication: Memory capacity of two layer neural networks with smooth activations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6583717)