← Search

Shira Kritchman

2 accepted papers

2020

Frequency Bias in Neural Networks for Input of Non-Uniform Density

ICML 2020poster

Recent works have partly attributed the generalization ability of over-parameterized neural networks to frequency bias – networks trained with gradient descent on data drawn from a uniform distribution find a low frequency fit before high frequency ones. As realistic training sets are not drawn from…

Cited by 220SourcePDFScholar
2019

The Convergence Rate of Neural Networks for Learned Functions of Different Frequencies

NeurIPS 2019poster

We study the relationship between the frequency of a function and the speed at which a neural network learns it. We build on recent results that show that the dynamics of overparameterized neural networks trained with gradient descent can be well approximated by a linear system. When normalized tr…