A spectral-based analysis of the separation between two-layer neural
networks and linear methods

v1v2 (latest)

A spectral-based analysis of the separation between two-layer neural networks and linear methods

Journal of machine learning research (JMLR), 2021

10 August 2021

ArXiv (abs)PDF HTML

Papers citing "A spectral-based analysis of the separation between two-layer neural networks and linear methods"

4 / 4 papers shown

Title
Analyzing the Role of Permutation Invariance in Linear Mode ConnectivityInternational Conference on Artificial Intelligence and Statistics (AISTATS), 2025 Keyao Zhan Puheng Li Lei Wu MoMe 213 1 0 13 Mar 2025
Fast training of accurate physics-informed neural networks without gradient descent Chinmay Datar Taniya Kapoor Abhishek Chandra Q. Sun Erik Lien Bolager Iryna Burak Anna Veselovska Massimo Fornasier Felix Dietrich 170 5 0 31 May 2024
$The $L^\infty$ Learnability of Reproducing Kernel Hilbert Spaces$ The $L^\infty$ Learnability of Reproducing Kernel Hilbert Spaces Hongrui Chen Jihao Long Lei Wu 107 0 0 05 Jun 2023
Learning Lipschitz Functions by GD-trained Shallow Overparameterized ReLU Neural Networks Ilja Kuzborskij Csaba Szepesvári 177 5 0 28 Dec 2022