paper-with-me

홈 › Papers

On the Similarity between the Laplace and Neural Tangent Kernels

2020-07-03 · NeurIPS 2020 12 · Amnon Geifman, Abhay Yadav, Yoni Kasten, Meirav Galun, David Jacobs, Ronen Basri

Recent theoretical work has shown that massively overparameterized neural networks are equivalent to kernel regressors that use Neural Tangent Kernels(NTK). Experiments show that these kernel methods perform similarly to real neural networks. Here we show that NTK for fully connected networks is closely related to the standard Laplace kernel. We show theoretically that for normalized data on the hypersphere both kernels have the same eigenfunctions and their eigenvalues decay polynomially at the same rate, implying that their Reproducing Kernel Hilbert Spaces (RKHS) include the same sets of functions. This means that both kernels give rise to classes of functions with the same smoothness properties. The two kernels differ for data off the hypersphere, but experiments indicate that when data is properly normalized these differences are not significant. Finally, we provide experiments on real data comparing NTK and the Laplace kernel, along with a larger class of{\gamma}-exponential kernels. We show that these perform almost identically. Our results suggest that much insight about neural networks can be obtained from analysis of the well-known Laplace kernel, which has a simple closed-form.

📄 PDF Abstract BibTeX arXiv:2007.01580

Code (1)

fwilliams/neural-splines pytorch

Methods 이 논문이 사용한 방법론

NTK 설명 없음

Similar Papers 제목 키워드 기반

An Empirical Analysis of the Laplace and Neural Tangent Kernels

2022-08-07 · Ronaldas Paulius Lencevicius

The neural tangent kernel is a kernel function defined over the parameter distribution of an infinite width neural network. Despite the impracticality of this limit, the neural tangent kernel has allowed for a more direc…

Deep Neural Tangent Kernel and Laplace Kernel Have the Same RKHS

2020-09-22 · ICLR 2021 1 · Lin Chen, Sheng Xu

We prove that the reproducing kernel Hilbert spaces (RKHS) of a deep neural tangent kernel and the Laplace kernel include the same set of functions, when both kernels are restricted to the sphere $\mathbb{S}^{d-1}$. Addi…

Spectral Analysis of the Neural Tangent Kernel for Deep Residual Networks

2021-04-07 · Yuval Belfer, Amnon Geifman, Meirav Galun, Ronen Basri

Deep residual network architectures have been shown to achieve superior accuracy over classical feed-forward networks, yet their success is still not fully understood. Focusing on massively over-parameterized, fully conn…

Stochastic Marginal Likelihood Gradients using Neural Tangent Kernels

2023-06-06 · Alexander Immer, Tycho F. A. van der Ouderaa, Mark van der Wilk, Gunnar Rätsch 외

Selecting hyperparameters in deep learning greatly impacts its effectiveness but requires manual effort and expertise. Recent works show that Bayesian model selection with Laplace approximations can allow to optimize suc…

Hyperparameter OptimizationModel Selection

Scalable Linearized Laplace Approximation via Surrogate Neural Kernel

2026-01-29 · Luis A. Ortega, Simón Rodríguez-Santana, Daniel Hernández-Lobato arxiv

We introduce a scalable method to approximate the kernel of the Linearized Laplace Approximation (LLA). For this, we use a surrogate deep neural network (DNN) that learns a compact feature representation whose inner prod…

Out-of-Distribution Detection