Neural Networks as Kernel Learners: The Silent Alignment Effect

v1v2 (latest)

Neural Networks as Kernel Learners: The Silent Alignment Effect

29 October 2021

Alexander B. Atanasov

Cengiz Pehlevan

ArXiv (abs)PDF HTML

Papers citing "Neural Networks as Kernel Learners: The Silent Alignment Effect"

13 / 63 papers shown

Title
What Can the Neural Tangent Kernel Tell Us About Adversarial Robustness? Nikolaos Tsilivis Julia Kempe AAML 98 20 0 11 Oct 2022
The Influence of Learning Rule on Representation Dynamics in Wide Neural Networks Blake Bordelon Cengiz Pehlevan 115 23 0 05 Oct 2022
On Kernel Regression with Data-Dependent Kernels James B. Simon BDL 58 3 0 04 Sep 2022
A view of mini-batch SGD via generating functions: conditions of convergence, phase transitions, benefit from negative momenta Maksim Velikanov Denis Kuznedelev Dmitry Yarotsky 75 8 0 22 Jun 2022
Limitations of the NTK for Understanding Generalization in Deep Learning Nikhil Vyas Yamini Bansal Preetum Nakkiran 116 34 0 20 Jun 2022
Spectral Bias Outside the Training Set for Deep Networks in the Kernel Regime Benjamin Bowman Guido Montúfar 82 15 0 06 Jun 2022
Self-Consistent Dynamical Field Theory of Kernel Evolution in Wide Neural Networks Blake Bordelon Cengiz Pehlevan MLT 87 85 0 19 May 2022
A Framework and Benchmark for Deep Batch Active Learning for Regression David Holzmüller Viktor Zaverkin Johannes Kastner Ingo Steinwart UQCV BDL GP 106 37 0 17 Mar 2022
Tight Convergence Rate Bounds for Optimization Under Power Law Spectral Conditions Maksim Velikanov Dmitry Yarotsky 97 8 0 02 Feb 2022
Separation of Scales and a Thermodynamic Description of Feature Learning in Some CNNs Inbar Seroussi Gadi Naveh Zohar Ringel 95 55 0 31 Dec 2021
Depth induces scale-averaging in overparameterized linear Bayesian neural networks Jacob A. Zavatone-Veth Cengiz Pehlevan BDL UQCV MDE 99 11 0 23 Nov 2021
A Theory of Neural Tangent Kernel Alignment and Its Influence on Training H. Shan Blake Bordelon 62 12 0 29 May 2021
Properties of the After Kernel Philip M. Long 66 29 0 21 May 2021