The curse of overparametrization in adversarial training: precise analysis of robust generalization for random features regression

Asymptotic properties of parametric estimators (62F12) Learning and adaptive systems in artificial intelligence (68T05) Robustness and adaptive procedures (parametric inference) (62F35) Neural nets and related approaches to inference from stochastic processes (62M45)

Recommendations

Precise statistical analysis of classification accuracies for adversarial training
The Generalization Error of Random Features Regression: Precise Asymptotics and the Double Descent Curve
Theoretical investigation of generalization bounds for adversarial learning of deep neural networks
Dimensionality Reduction, Regularization, and Generalization in Overparameterized Regressions
Adversarial examples in random neural networks with general activations

Cites work

scientific article; zbMATH DE number 4061904 (Why is no real title available?)
A mean field view of the landscape of two-layer neural networks
A model of double descent for high-dimensional binary linear classification
A precise high-dimensional asymptotic theory for boosting and minimum-$\ell_1$-norm interpolated classifiers
A random matrix approach to neural networks
Analysis of a two-layer neural network via displacement convexity
Deep learning: a statistical viewpoint
DeepMoM: Robust Deep Learning With Median-of-Means
Eine neue Herleitung des Exponentialgesetzes in der Wahrscheinlichkeitsrechnung.
Modern Coding Theory
Nonlinear random matrix theory for deep learning
On the Adversarial Robustness of Robust Estimators
On the impact of predictor geometry on the performance on high-dimensional ridge-regularized generalized robust regression estimators
On the robustness to adversarial corruption and to heavy-tailed data of the Stahel–Donoho median of means
Precise Error Analysis of Regularized <inline-formula> <tex-math notation="LaTeX">$M$ </tex-math> </inline-formula>-Estimators in High Dimensions
Precise statistical analysis of classification accuracies for adversarial training
Provable tradeoffs in adversarially robust classification
Reconciling modern machine-learning practice and the classical bias-variance trade-off
Surprises in high-dimensional ridgeless least squares interpolation
The Generalization Error of Random Features Regression: Precise Asymptotics and the Double Descent Curve
The implicit bias of gradient descent on separable data
The spectrum of random inner-product kernel matrices
Theoretical Insights Into the Optimization Landscape of Over-Parameterized Shallow Neural Networks
Universality Laws for High-Dimensional Learning With Random Features

Cited in

(2)

Adversarial examples in random neural networks with general activations
Adversarial robustness for latent models: revisiting the robust-standard accuracies tradeoff

This page was built for publication: The curse of overparametrization in adversarial training: precise analysis of robust generalization for random features regression

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6550964)