On the Inductive Bias of Neural Tangent Kernels - Inria - Institut national de recherche en sciences et technologies du numérique
Conference Papers Year : 2019

On the Inductive Bias of Neural Tangent Kernels

Abstract

State-of-the-art neural networks are heavily over-parameterized, making the optimization algorithm a crucial ingredient for learning predictive models with good generalization properties. A recent line of work has shown that in a certain over-parameterized regime, the learning dynamics of gradient descent are governed by a certain kernel obtained at initialization, called the neural tangent kernel. We study the inductive bias of learning in such a regime by analyzing this kernel and the corresponding function space (RKHS). In particular, we study smoothness, approximation, and stability properties of functions with finite norm, including stability to image deformations in the case of convolutional networks, and compare to other known kernels for similar architectures.
Fichier principal
Vignette du fichier
main.pdf (363.7 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

hal-02144221 , version 1 (29-05-2019)
hal-02144221 , version 2 (25-10-2019)

Identifiers

Cite

Alberto Bietti, Julien Mairal. On the Inductive Bias of Neural Tangent Kernels. NeurIPS 2019 - Thirty-third Conference on Neural Information Processing Systems, Dec 2019, Vancouver, Canada. pp.12873-12884. ⟨hal-02144221v2⟩
403 View
1064 Download

Altmetric

Share

More