Locality defeats the curse of dimensionality in convolutional teacher–student scenarios*

DOI10.1088/1742-5468/AC98ABMaRDI QIDQ5055428zbMATH OpenOpenAlexFDO

Authors Alessandro Favero, Francesco Cagnetta, Matthieu Wyart

Publication date 13 December 2022

Published in Journal of Statistical Mechanics: Theory and Experiment (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://arxiv.org/abs/2106.08619

zbMATH Keywords

deep learninglearning theorymachine learninganalysis of algorithms

Mathematics Subject Classification ID

Statistical mechanics, structure of matter (82-XX)

Recommendations

• Asymptotic learning curves of kernel methods: empirical data versus teacher–student paradigm
• Learning curves of generic features maps for realistic datasets with a teacher-student model*
• Just interpolate: kernel ``ridgeless regression can generalize
• When do neural networks outperform kernel methods?*
• VC dimensions of group convolutional neural networks

Cites work

• scientific article; zbMATH DE number 1273988 (Why is no real title available?)
• Asymptotic behavior of the eigenvalues of certain integral equations. II
• Asymptotic learning curves of kernel methods: empirical data versus teacher–student paradigm
• Bayesian learning for neural networks
• Breaking the curse of dimensionality with convex neural networks
• Distance-based classification with Lipschitz functions
• ImageNet
• Theoretical issues in deep networks
• Wide neural networks of any depth evolve as linear models under gradient descent ^*

Cited in

(1)

• Learning curves of generic features maps for realistic datasets with a teacher-student model*

Describes a project that uses

Uses Software

• ImageNet

This page was built for publication: Locality defeats the curse of dimensionality in convolutional teacher–student scenarios*

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5055428)