Deep regularization and direct training of the inner layers of neural networks with kernel flows

DOI10.1016/J.PHYSD.2021.132952MaRDI QIDQ2077570zbMATH OpenOpenAlexFDO

Authors Gene Ryan Yoo, Houman Owhadi

Publication date 21 February 2022

Published in Physica D (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2002.08335, https://authors.library.caltech.edu/106580/

zbMATH Keywords

artificial neural networksGaussian process regressionimage classificationinner layer trainingkernel flowsmachine learning

Mathematics Subject Classification ID

Artificial intelligence (68Txx)

Recommendations

• Kernel flows: from learning kernels from data into the abyss
• Regularization via mass transportation
• Stochastic training of residual networks in deep learning
• Kernel analysis of deep networks
• On Kernel Method–Based Connectionist Models and Supervised Deep Learning Without Backpropagation

Cites work

• A comparison of generalized cross validation and modified maximum likelihood for estimating the parameters of a stochastic process
• Bayesian Inference for Non-Stationary Spatial Covariance Structure via Spatial Deformations
• Bayesian numerical homogenization
• Can a finite element method perform arbitrarily badly?
• Cross validation and maximum likelihood estimations of hyper-parameters of Gaussian processes with model misspecification
• scientific article; zbMATH DE number 6378127 (Why is no real title available?)
• Kernel flows: from learning kernels from data into the abyss
• Operator-adapted wavelets, fast solvers, and numerical homogenization. From a game theoretic approach to numerical approximation and algorithm design
• Prevalence of neural collapse during the terminal phase of deep learning training
• Statistical Numerical Approximation

Cited in

(11)

• Learning ``best kernels from data in Gaussian process regression. With application to aerodynamics
• Do ideas have shape? Idea registration as the continuous limit of artificial neural networks
• Kernel flows: from learning kernels from data into the abyss
• Computational graph completion
• Learning dynamical systems from data: a simple cross-validation perspective. III: Irregularly-sampled time series
• On the regularization of convolutional kernel tensors in neural networks
• On Kernel Method–Based Connectionist Models and Supervised Deep Learning Without Backpropagation
• scientific article; zbMATH DE number 7415108 (Why is no real title available?)
• Learning dynamical systems from data: a simple cross-validation perspective. V: Sparse kernel flows for 132 chaotic dynamical systems
• Hausdorff metric based training of kernels to learn attractors with application to 133 chaotic dynamical systems
• Kernel sum of squares for data adapted kernel learning of dynamical systems from data: a global optimization approach

Describes a project that uses

Uses Software

• TensorFlow
• BranchyNet
• GitHub

This page was built for publication: Deep regularization and direct training of the inner layers of neural networks with kernel flows

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2077570)