SGD Learns Over-parameterized Networks that Provably Generalize on Linearly Separable Data

27 October 2017

Papers citing "SGD Learns Over-parameterized Networks that Provably Generalize on Linearly Separable Data"

2 / 52 papers shown

Title
Gradient descent with identity initialization efficiently learns positive definite linear transformations by deep residual networks Peter L. Bartlett D. Helmbold Philip M. Long 23 116 0 16 Feb 2018
Norm-Based Capacity Control in Neural Networks Behnam Neyshabur Ryota Tomioka Nathan Srebro 119 577 0 27 Feb 2015