mlsanityData to 14 Oct 2026

InfoAT: Improving Adversarial Training Using the Information Bottleneck Principle

arXiv 2206.12292 ↗First posted 23 Jun 2022cs.LG cs.CVcited by 7

Not on the map. Why?

Abstract (arXiv)

Adversarial training (AT) has shown excellent high performance in defending against adversarial examples. Recent studies demonstrate that examples are not equally important to the final robustness of models during AT, that is, the so-called hard examples that can be attacked easily exhibit more influence than robust examples on the final robustness. Therefore, guaranteeing the robustness of hard examples is crucial for improving the final robustness of the model. However, defining effective heuristics to search for hard examples is still difficult. In this article, inspired by the information bottleneck (IB) principle, we uncover that an example with high mutual information of the input and its associated latent representation is more likely to be attacked. Based on this observation, we propose a novel and effective adversarial training method (InfoAT). InfoAT is encouraged to find examples with high mutual information and exploit them efficiently to improve the final robustness of models. Experimental results show that InfoAT achieves the best robustness among different datasets and models in comparison with several state-of-the-art methods.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 33 of its 55 references that are in the map. First posted on arXiv 23 Jun 2022.

Builds on

33 of 55 references.

References · 33

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  3. 2013Intriguing properties of neural networkscited by 5,155
  4. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,517
  5. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,451
  6. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,268
  7. 2019Adversarial Examples Are Not Bugs, They Are Featurescited by 953
  8. 2019Square Attack: a query-efficient black-box adversarial attack via random searchcited by 685
  9. 2018MINE: Mutual Information Neural Estimationcited by 683
  10. 2020On Adaptive Attacks to Adversarial Example Defensescited by 555
  11. 2017Countering Adversarial Images using Input Transformationscited by 520
  12. 2019Unlabeled Data Improves Adversarial Robustnesscited by 416
21 more
  1. 2019Minimally distorted Adversarial Examples with a Fast Adaptive Boundary Attackcited by 334
  2. 2017MagNet: a Two-Pronged Defense against Adversarial Examplescited by 269
  3. 2015DeepDriving: Learning Affordance for Direct Perception in Autonomous Drivingcited by 231
  4. 2018Adversarial Logit Pairingcited by 224
  5. 2018Adversarial Risk and the Dangers of Evaluating Against Weak Attackscited by 221
  6. 2019Towards Stable and Efficient Training of Verifiably Robust Neural Networkscited by 205
  7. 2020Geometry-aware Instance-reweighted Adversarial Trainingcited by 198
  8. 2019Are Labels Required for Improving Adversarial Robustness?cited by 163
  9. 2019Understanding the Limitations of Variational Mutual Information Estimatorscited by 140
  10. 2016Information Dropout: Learning Optimal Representations Through Noisy Computationcited by 132
  11. 2020Maximum-Entropy Adversarial Data Augmentation for Improved Generalization and Robustnesscited by 121
  12. 2020RayS: A Ray Searching Method for Hard-label Adversarial Attackcited by 96
  13. 2019Defense Against Adversarial Attacks Using Feature Scattering-based Adversarial Trainingcited by 95
  14. 2018Learning Representations for Neural Network-Based Classification Using the Information Bottleneck Principlecited by 70
  15. 2018Uncertainty in the Variational Information Bottleneckcited by 39
  16. 2018Adversarial Defense by Stratified Convolutional Sparse Codingcited by 31
  17. 2021Towards Evaluating the Robustness of Deep Diagnostic Models by Adversarial Attackcited by 15
  18. 2021Towards Evaluating and Training Verifiably Robust Neural Networkscited by 15
  19. 2016Measuring multivariate redundant information with pointwise common change in surprisalcited by 13
  20. 2019GanDef: A GAN based Adversarial Training Defense for Neural Network Classifiercited by 6
  21. 2019ZK-GanDef: A GAN based Zero Knowledge Adversarial Training Defense for Neural Networkscited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 1; 2023: 3; 2024: 0; 2025: 1; 2026: 2.

7 papers.

  1. 2023Survey on Adversarial Attack and Defense for Medical Image Analysis: Methods and Challengescited by 17
  2. 2023Defenses in Adversarial Machine Learning: A Surveycited by 14
  3. 2023Improved Adversarial Training Through Adaptive Instance-wise Loss Smoothingcited by 5
  4. 2025Weakly Supervised Contrastive Adversarial Training for Learning Robust Features from Semi-supervised Datacited by 1
  5. 2022Robust Few-shot Learning Without Using any Adversarial Samplescited by 0
  6. 2026GeoIB: Geometry-Aware Information Bottleneck via Statistical-Manifold Compressioncited by 0
  7. 2026Improving the Robustness/Accuracy Tradeoff Against Adversarial Attacks Using Information Bottleneck Distillation Through Dual Teacherscited by 0