mlsanityData to 7 Oct 2026

Maintaining Adversarial Robustness in Continuous Learning

arXiv 2402.11196 ↗First posted 17 Feb 2024cs.LG cs.AIcited by 5

Not on the map. Why?

Abstract (arXiv)

Adversarial robustness is essential for security and reliability of machine learning systems. However, adversarial robustness enhanced by defense algorithms is easily erased as the neural network's weights update to learn new tasks. To address this vulnerability, it is essential to improve the capability of neural networks in terms of robust continual learning. Specially, we propose a novel gradient projection technique that effectively stabilizes sample gradients from previous data by orthogonally projecting back-propagation gradients onto a crucial subspace before using them for weight updates. This technique can maintaining robustness by collaborating with a class of defense algorithms through sample gradient smoothing. The experimental results on four benchmarks including Split-CIFAR100 and Split-miniImageNet, demonstrate that the superiority of the proposed approach in mitigating rapidly degradation of robustness during continual learning even when facing strong adversarial attacks.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 25 of its 32 references that are in the map. First posted on arXiv 17 Feb 2024.

Builds on

25 of 32 references.

References · 25

  1. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  3. 2016Overcoming catastrophic forgetting in neural networkscited by 4,718
  4. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  5. 2017Gradient Episodic Memory for Continual Learningcited by 1,871
  6. 2016Adversarial examples in the physical worldcited by 1,762
  7. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  8. 2018Efficient Lifelong Learning with A-GEMcited by 1,044
  9. 2023A Comprehensive Survey of Continual Learning: Theory, Method and Applicationcited by 860
  10. 2013An Empirical Investigation of Catastrophic Forgetting in Gradient-Based Neural Networkscited by 711
  11. 2015Distillation as a Defense to Adversarial Perturbations against Deep Neural Networkscited by 675
  12. 2019Orthogonal Gradient Descent for Continual Learningcited by 361
13 more
  1. 2021Gradient Projection Memory for Continual Learningcited by 338
  2. 2019On Tiny Episodic Memories in Continual Learningcited by 299
  3. 2017Parseval Networks: Improving Robustness to Adversarial Examplescited by 288
  4. 2017Improving the Adversarial Robustness and Interpretability of Deep Neural Networks by Regularizing their Input Gradientscited by 236
  5. 2018Continual Learning of Context-dependent Processing in Neural Networkscited by 178
  6. 2021Training Networks in Null Space of Feature Covariance for Continual Learningcited by 139
  7. 2021Avalanche: an End-to-End Library for Continual Learningcited by 124
  8. 2016Robust Large Margin Deep Neural Networkscited by 117
  9. 2020Continual Learning in Low-rank Orthogonal Subspacescited by 99
  10. 2020Opportunities and Challenges in Deep Learning Adversarial Robustness: A Surveycited by 71
  11. 2021Flattening Sharpness for Dynamic Gradient Projection Memory Benefits Continual Learningcited by 71
  12. 2022Continual Learning with Recursive Gradient Optimizationcited by 32
  13. 2023Towards Adversarially Robust Continual Learningcited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 4; 2026: 1.

5 papers.

  1. 2025From Large AI Models to Agentic AI: A Tutorial on Future Intelligent Communicationscited by 1
  2. 2025Noise-Tolerant Coreset-Based Class Incremental Continual Learningcited by 0
  3. 2025SHIELD: Secure Hypernetworks for Incremental Expansion Learning Defensecited by 0
  4. 2025Exploring Cross-Stage Adversarial Transferability in Class-Incremental Continual Learningcited by 0
  5. 2026Repurposing Adversarial Perturbations for Continual Learning: From Defense to Active Alignmentcited by 0