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

DOI10.1088/1742-5468/ac98abOpenAlexW3213007444MaRDI QIDQ5055428

Francesco Cagnetta, Alessandro Favero, Matthieu Wyart

Publication date: 13 December 2022

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

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

zbMATH Keywords

learning theoryanalysis of algorithmsmachine learningdeep learning

Mathematics Subject Classification ID

Statistical mechanics, structure of matter (82-XX)

Uses Software

• ImageNet

Cites Work

• Unnamed Item
• Unnamed Item
• ImageNet
• Bayesian learning for neural networks
• Asymptotic behavior of the eigenvalues of certain integral equations. II
• Theoretical issues in deep networks
• Breaking the Curse of Dimensionality with Convex Neural Networks
• Asymptotic learning curves of kernel methods: empirical data versus teacher–student paradigm
• Wide neural networks of any depth evolve as linear models under gradient descent ^*