Deterministic equivalent and error universality of deep random features learning

DOI10.1088/1742-5468/AD65E2MaRDI QIDQ7032695zbMATH OpenFDO

Authors Dominik Schröder, Hugo Cui, Daniil Dmitriev, Bruno Loureiro

Publication date 4 February 2025

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://doi.org/10.1088/1742-5468/AD65E2

zbMATH Keywords

deep learninglearning theorymachine learning

Mathematics Subject Classification ID

Statistical mechanics, structure of matter (82-XX)

Cites work

• A note on the Hanson-Wright inequality for random vectors with dependencies
• A precise high-dimensional asymptotic theory for boosting and minimum-\(\ell_1\)-norm interpolated classifiers
• Anisotropic local laws for random matrices
• DISTRIBUTION OF EIGENVALUES FOR SOME SETS OF RANDOM MATRICES
• Eigenvalue distribution of some nonlinear models of random matrices
• Generalization error of random feature and kernel methods: hypercontractivity and kernel matrix concentration
• High-dimensional asymptotics of prediction: ridge regression and classification
• High-dimensional probability. An introduction with applications in data science
• scientific article; zbMATH DE number 5278585 (Why is no real title available?)
• Learning curves of generic features maps for realistic datasets with a teacher-student model*
• Nonlinear random matrix theory for deep learning
• On a surprising relation between the Marchenko-Pastur law, rectangular and square free convolutions
• On robust regression with high-dimensional predictors
• Spectrum estimation for large dimensional covariance matrices using random matrix theory
• Surprises in high-dimensional ridgeless least squares interpolation
• The generalised product moment distribution in samples from a normal multivariate population.
• The Generalization Error of Random Features Regression: Precise Asymptotics and the Double Descent Curve
• Triple descent and the two kinds of overfitting: where and why do they appear?*
• When do neural networks outperform kernel methods?*

Cited in

(1)

• Bayes-optimal learning of deep random networks of extensive-width

This page was built for publication: Deterministic equivalent and error universality of deep random features learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7032695)