Adversarial Training with Generated Data in High-Dimensional Regression: An Asymptotic Study
Not on the map. Why?
Abstract (arXiv)
In recent years, studies such as citecarmon2019unlabeled,gowal2021improving,xing2022artificial have demonstrated that incorporating additional real or generated data with pseudo-labels can enhance adversarial training through a two-stage training approach. In this paper, we perform a theoretical analysis of the asymptotic behavior of this method in high-dimensional linear regression. While a double-descent phenomenon can be observed in ridgeless training, with an appropriate L2 regularization, the two-stage adversarial training achieves a better performance. Finally, we derive a shortcut cross-validation formula specifically tailored for the two-stage training method.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 31 of its 44 references that are in the map. First posted on arXiv 21 Jun 2023.
Builds on
31 of 44 references.
References · 31
- 2014Explaining and Harnessing Adversarial Examplescited by 7,364
- 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,517
- 2020RobustBench: a standardized adversarial robustness benchmarkcited by 661
- 2019Surprises in High-Dimensional Ridgeless Least Squares Interpolationcited by 513
- 2019Unlabeled Data Improves Adversarial Robustnesscited by 416
- 2019Using Pre-Training Can Improve Model Robustness and Uncertaintycited by 373
- 2017Certifying Some Distributional Robustness with Principled Adversarial Trainingcited by 337
- 2020Attacks Which Do Not Kill Training Make Adversarial Learning Strongercited by 243
- 2021Improving Robustness using Generated Datacited by 234
- 2023Better Diffusion Models Further Improve Adversarial Trainingcited by 208
- 2019Two models of double descent for weak featurescited by 195
- 2018Rademacher Complexity for Adversarially Robust Generalizationcited by 132
19 more
- 2019Adversarial Training Can Hurt Generalizationcited by 132
- 2021On the Convergence and Robustness of Adversarial Trainingcited by 130
- 2020Double Trouble in Double Descent : Bias and Variance(s) in the Lazy Regimecited by 95
- 2018Curriculum Adversarial Trainingcited by 89
- 2020Precise Tradeoffs in Adversarial Training for Linear Regressioncited by 77
- 2020Understanding Double Descent Requires a Fine-Grained Bias-Variance Decompositioncited by 76
- 2019Robustness to Adversarial Perturbations in Learning from Incomplete Datacited by 49
- 2020The Curious Case of Adversarially Robust Models: More Data Can Help, Double Descend, or Hurt Generalizationcited by 39
- 2020Kernel regression in high dimensions: Refined analysis beyond double descentcited by 39
- 2020Over-parameterized Adversarial Training: An Analysis Overcoming the Curse of Dimensionalitycited by 36
- 2022Stability Analysis and Generalization Bounds of Adversarial Trainingcited by 30
- 2020Sharp Statistical Guarantees for Adversarially Robust Gaussian Classificationcited by 28
- 2020More Data Can Expand the Generalization Gap Between Adversarially Robust and Standard Modelscited by 27
- 2020Memorizing without overfitting: Bias, variance, and interpolation in over-parameterized modelscited by 27
- 2020Statistical Guarantees for Regularized Neural Networkscited by 25
- 2020Improving Adversarial Robustness via Unlabeled Out-of-Domain Datacited by 23
- 2020Adversarially Robust Estimate and Risk Analysis in Linear Regressioncited by 22
- 2022Adversarial Rademacher Complexity of Deep Neural Networkscited by 18
- 2023Understanding Overfitting in Adversarial Training via Kernel Regressioncited by 2
Built on by
Citing papers per year: 2025: 1.
1 paper.