Neural Networks Learning and Memorization with (almost) no Over-Parameterization

22 November 2019

Papers citing "Neural Networks Learning and Memorization with (almost) no Over-Parameterization"

10 / 10 papers shown

Title
Analysis of the expected $L_2$ error of an over-parametrized deep neural network estimate learned by gradient descent without regularization Selina Drews Michael Kohler 36 2 0 24 Nov 2023
Efficient SGD Neural Network Training via Sublinear Activated Neuron Identification Lianke Qin Zhao Song Yuanyuan Yang 25 9 0 13 Jul 2023
Sharp Lower Bounds on Interpolation by Deep ReLU Neural Networks at Irregularly Spaced Data Jonathan W. Siegel 14 2 0 02 Feb 2023
When Expressivity Meets Trainability: Fewer than $n$ Neurons Can Work Jiawei Zhang Yushun Zhang Mingyi Hong Ruoyu Sun Zhi-Quan Luo 26 10 0 21 Oct 2022
Size and depth of monotone neural networks: interpolation and approximation Dan Mikulincer Daniel Reichman 28 7 0 12 Jul 2022
Bounding the Width of Neural Networks via Coupled Initialization -- A Worst Case Analysis Alexander Munteanu Simon Omlor Zhao Song David P. Woodruff 30 15 0 26 Jun 2022
Randomly Initialized One-Layer Neural Networks Make Data Linearly Separable Promit Ghosal Srinath Mahankali Yihang Sun MLT 26 4 0 24 May 2022
Subquadratic Overparameterization for Shallow Neural Networks Chaehwan Song Ali Ramezani-Kebrya Thomas Pethick Armin Eftekhari V. Cevher 27 31 0 02 Nov 2021
The Interpolation Phase Transition in Neural Networks: Memorization and Generalization under Lazy Training Andrea Montanari Yiqiao Zhong 47 95 0 25 Jul 2020
Learning Parities with Neural Networks Amit Daniely Eran Malach 24 76 0 18 Feb 2020