English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In this paper, we study the generalization performance of min $\ell_2$-normoverfitting solutions for the neural tangent kernel (NTK) model of a two-layerneural network with ReLU activation that has no bias term. We show that,depending on the ground-truth function, the test error of overfitted NTK modelsexhibits characteristics that are different from the "double-descent" of otheroverparameterized linear models with simple Fourier or Gaussian features.Specifically, for a class of learnable functions, we provide a new upper boundof the generalization error that approaches a small limiting value, even whenthe number of neurons $p$ approaches infinity. This limiting value furtherdecreases with the number of training samples $n$. For functions outside ofthis class, we provide a lower bound on the generalization error that does notdiminish to zero even when $n$ and $p$ are both large.

On the Generalization Power of Overfitted Two-Layer Neural Tangent Kernel Models