Timezone: »

 
Poster
On the Inductive Bias of Neural Tangent Kernels
Alberto Bietti · Julien Mairal

Thu Dec 12 05:00 PM -- 07:00 PM (PST) @ East Exhibition Hall B + C #245
in Theory -- Spaces of Functions and Kernels »

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.

Keywords: Regularization · Spaces of Functions and Kernels · Theory · Algorithms -> Kernel Methods; Theory
[ Paper]  [ Poster] 

Author Information

Alberto Bietti (Inria)
Julien Mairal (Inria)

More from the Same Authors