mlsanityData to 7 Oct 2026

Interpolated Adversarial Training: Achieving Robust Neural Networks without Sacrificing Too Much Accuracy

arXiv 1906.06784 ↗First posted 16 Jun 2019stat.ML cs.LGcited by 49

Not on the map. Why?

Abstract (arXiv)

Adversarial robustness has become a central goal in deep learning, both in the theory and the practice. However, successful methods to improve the adversarial robustness (such as adversarial training) greatly hurt generalization performance on the unperturbed data. This could have a major impact on how the adversarial robustness affects real world systems (i.e. many may opt to forego robustness if it can improve accuracy on the unperturbed data). We propose Interpolated Adversarial Training, which employs recently proposed interpolation based training methods in the framework of adversarial training. On CIFAR-10, adversarial training increases the standard test error (when there is no adversary) from 4.43% to 12.32%, whereas with our Interpolated adversarial training we retain the adversarial robustness while achieving a standard test error of only 6.45%. With our technique, the relative increase in the standard error for the robust model is reduced from 178.1% to just 45.5%. Moreover, we provide mathematical analysis of Interpolated Adversarial Training to confirm its efficiencies and demonstrate its advantages in terms of robustness and generalization.

Key facts. Cited by 49 papers in the map (data through 7 Oct 2026). First posted on arXiv 16 Jun 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 14; 2022: 17; 2023: 10; 2024: 4; 2025: 3; 2026: 1.

49 papers.

Sort:

  1. 2021Evaluating Gradient Inversion Attacks and Defenses in Federated Learningcited by 196
  2. 2021Admix: Enhancing the Transferability of Adversarial Attackscited by 155
  3. 2021Relating Adversarially Robust Generalization to Flat Minimacited by 57
  4. 2021SmoothMix: Training Confidence-calibrated Smoothed Classifiers for Certified Robustnesscited by 49
  5. 2022A Survey of Mix-based Data Augmentation: Taxonomy, Methods, Applications, and Explainabilitycited by 30
  6. 2021Noisy Feature Mixupcited by 27
  7. 2021Survey: Image Mixing and Deleting for Data Augmentationcited by 25
  8. 2022Enhance the Visual Representation via Discrete Adversarial Trainingcited by 23
  9. 2021CIFS: Improving Adversarial Robustness of CNNs via Channel-wise Importance-based Feature Selectioncited by 21
  10. 2021Towards Understanding the Data Dependency of Mixup-style Trainingcited by 19
  11. 2023Improving the Accuracy-Robustness Trade-Off of Classifiers via Adaptive Smoothingcited by 18
  12. 2024A Survey on Mixup Augmentations and Beyondcited by 18
37 more
  1. 2022NoisyMix: Boosting Model Robustness to Common Corruptionscited by 17
  2. 2024MixedNUTS: Training-Free Accuracy-Robustness Balance via Nonlinearly Mixed Classifierscited by 16
  3. 2022Provably Learning Diverse Features in Multi-View Data with Midpoint Mixupcited by 14
  4. 2023Generalist: Decoupling Natural and Robust Generalizationcited by 14
  5. 2022Improving Robustness by Enhancing Weak Subnetscited by 12
  6. 2022Why adversarial training can hurt robust accuracycited by 12
  7. 2021Interpolated Joint Space Adversarial Training for Robust and Generalizable Defensescited by 10
  8. 2021Guided Interpolation for Adversarial Trainingcited by 8
  9. 2021Adversarially Optimized Mixup for Robust Classificationcited by 7
  10. 2022Revisiting the Adversarial Robustness-Accuracy Tradeoff in Robot Learningcited by 6
  11. 2022Modality-Balanced Embedding for Video Retrievalcited by 6
  12. 2022Towards Adversarial Purification using Denoising AutoEncoderscited by 6
  13. 2022MixupE: Understanding and Improving Mixup from Directional Derivative Perspectivecited by 6
  14. 2024Structure-Preserving Physics-Informed Neural Networks With Energy or Lyapunov Structurecited by 6
  15. 2022Learn2Weight: Parameter Adaptation against Similar-domain Adversarial Attackscited by 5
  16. 2025Optimal Transport-Guided Source-Free Adaptation for Face Anti-Spoofingcited by 5
  17. 2021MixSiam: A Mixture-based Approach to Self-supervised Representation Learningcited by 4
  18. 2022Adversarial robustness of VAEs through the lens of local geometrycited by 4
  19. 2023Exploring Adversarial Attacks on Neural Networks: An Explainable Approachcited by 4
  20. 2021Adaptive Feature Alignment for Adversarial Trainingcited by 3
  21. 2023Robust Feature Inference: A Test-time Defense Strategy using Spectral Projectionscited by 3
  22. 2022Universum-inspired Supervised Contrastive Learningcited by 2
  23. 2022Robust Sensible Adversarial Learning of Deep Neural Networks for Image Classificationcited by 2
  24. 2022Data-free Defense of Black Box Models Against Adversarial Attackscited by 2
  25. 2023NoiseCAM: Explainable AI for the Boundary Between Noise and Adversarial Attackscited by 2
  26. 2023Mixing Classifiers to Alleviate the Accuracy-Robustness Trade-Offcited by 2
  27. 2022AugRmixAT: A Data Processing and Training Method for Improving Multiple Robustness and Generalization Performancecited by 1
  28. 2023Towards Brain Inspired Design for Addressing the Shortcomings of ANNscited by 1
  29. 2025Generalist++: A Meta-learning Framework for Mitigating Trade-off in Adversarial Trainingcited by 1
  30. 2021Improving Transformation-based Defenses against Adversarial Examples with First-order Perturbationscited by 0
  31. 2022Unrestricted Adversarial Samples Based on Non-semantic Feature Clusters Substitutioncited by 0
  32. 2023[Re] Double Sampling Randomized Smoothingcited by 0
  33. 2023G-Mix: A Generalized Mixup Learning Framework Towards Flat Minimacited by 0
  34. 2023On the Equivalence of Graph Convolution and Mixupcited by 0
  35. 2024Adversarial Training in Low-Label Regimes with Margin-Based Interpolationcited by 0
  36. 2025Towards Adversarially Robust Deep Metric Learningcited by 0
  37. 2026Zero-Sacrifice Persistent-Robustness Adversarial Defense for Pre-Trained Encoderscited by 0