mlsanityData to 7 Oct 2026

Beckman Defense

arXiv 2301.01495 ↗First posted 4 Jan 2023cs.LG cs.CR cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Optimal transport (OT) based distributional robust optimisation (DRO) has received some traction in the recent past. However, it is at a nascent stage but has a sound potential in robustifying the deep learning models. Interestingly, OT barycenters demonstrate a good robustness against adversarial attacks. Owing to the computationally expensive nature of OT barycenters, they have not been investigated under DRO framework. In this work, we propose a new barycenter, namely Beckman barycenter, which can be computed efficiently and used for training the network to defend against adversarial attacks in conjunction with adversarial training. We propose a novel formulation of Beckman barycenter and analytically obtain the barycenter using the marginals of the input image. We show that the Beckman barycenter can be used to train adversarially trained networks to improve the robustness. Our training is extremely efficient as it requires only a single epoch of training. Elaborate experiments on CIFAR-10, CIFAR-100 and Tiny ImageNet demonstrate that training an adversarially robust network with Beckman barycenter can significantly increase the performance. Under auto attack, we get a a maximum boost of 10% in CIFAR-10, 8.34% in CIFAR-100 and 11.51% in Tiny ImageNet. Our code is available at https://github.com/Visual-Conception-Group/test-barycentric-defense.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 40 of its 71 references that are in the map. First posted on arXiv 4 Jan 2023.

Builds on

40 of 71 references.

References · 40

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  3. 2013Intriguing properties of neural networkscited by 5,114
  4. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  5. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,277
  6. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  7. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  8. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  9. 2016Adversarial Machine Learning at Scalecited by 934
  10. 2015The Limitations of Deep Learning in Adversarial Settingscited by 784
  11. 2020Fast is better than free: Revisiting adversarial trainingcited by 723
  12. 2018Improving Transferability of Adversarial Examples with Input Diversitycited by 591
28 more
  1. 2022Diffusion Models for Adversarial Purificationcited by 469
  2. 2018Invariant Information Clustering for Unsupervised Image Classification and Segmentationcited by 454
  3. 2019Unlabeled Data Improves Adversarial Robustnesscited by 415
  4. 2021Recent Advances in Adversarial Training for Adversarial Robustnesscited by 276
  5. 2020Attacks Which Do Not Kill Training Make Adversarial Learning Strongercited by 242
  6. 2018Adversarial Logit Pairingcited by 224
  7. 2013Fast Computation of Wasserstein Barycenterscited by 216
  8. 2021Fixing Data Augmentation to Improve Adversarial Robustnesscited by 201
  9. 2016Statistics of Robust Optimization: A Generalized Empirical Likelihood Approachcited by 182
  10. 2018Understanding Measures of Uncertainty for Adversarial Example Detectioncited by 177
  11. 2021Adversarial purification with Score-based generative modelscited by 132
  12. 2021On the Convergence and Robustness of Adversarial Trainingcited by 127
  13. 2022LAS-AT: Adversarial Training with Learnable Attack Strategycited by 94
  14. 2020Learnable Boundary Guided Adversarial Trainingcited by 93
  15. 2018Curriculum Adversarial Trainingcited by 89
  16. 2019Wasserstein Adversarial Examples via Projected Sinkhorn Iterationscited by 89
  17. 2020CAT: Customized Adversarial Training for Improved Robustnesscited by 61
  18. 2021Adversarial Attacks are Reversible with Natural Supervisioncited by 47
  19. 2021Attacking Adversarial Attacks as A Defensecited by 31
  20. 2015A Smoothed Dual Approach for Variational Wasserstein Problemscited by 29
  21. 2022Robust Weight Perturbation for Adversarial Trainingcited by 25
  22. 2020Principled learning method for Wasserstein distributionally robust optimization with local perturbationscited by 14
  23. 2021On the complexity of the optimal transport problem with graph-structured costcited by 14
  24. 2022A2: Efficient Automated Attacker for Boosting Adversarial Trainingcited by 10
  25. 2021Learning to Generate Wasserstein Barycenterscited by 6
  26. 2018Solving Large-Scale Optimization Problems with a Convergence Rate Independent of Grid Sizecited by 5
  27. 2017Fast GPU-Based Seismogram Simulation from Microseismic Events in Marine Environments Using Heterogeneous Velocity Modelscited by 4
  28. 2021Internal Wasserstein Distance for Adversarial Attack and Defensecited by 1

Built on by

Not cited yet.