mlsanityData to 7 Oct 2026

Adversarial Robustness by Design through Analog Computing and Synthetic Gradients

arXiv 2101.02115 ↗First posted 6 Jan 2021cs.CV cs.LGcited by 7

Not on the map. Why?

Abstract (arXiv)

We propose a new defense mechanism against adversarial attacks inspired by an optical co-processor, providing robustness without compromising natural accuracy in both white-box and black-box settings. This hardware co-processor performs a nonlinear fixed random transformation, where the parameters are unknown and impossible to retrieve with sufficient precision for large enough dimensions. In the white-box setting, our defense works by obfuscating the parameters of the random projection. Unlike other defenses relying on obfuscated gradients, we find we are unable to build a reliable backward differentiable approximation for obfuscated parameters. Moreover, while our model reaches a good natural accuracy with a hybrid backpropagation - synthetic gradient method, the same approach is suboptimal if employed to generate adversarial examples. We find the combination of a random projection and binarization in the optical system also improves robustness against various types of black-box attacks. Finally, our hybrid training method builds robust features against transfer attacks. We demonstrate our approach on a VGG-like architecture, placing the defense on top of the convolutional features, on CIFAR-10 and CIFAR-100. Code is available at https://github.com/lightonai/adversarial-robustness-by-design.

Key facts. Cited by 7 papers in the map (data through 7 Oct 2026). Builds on 28 of its 32 references that are in the map. First posted on arXiv 6 Jan 2021.

Builds on

28 of 32 references.

References · 28

  1. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  3. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  4. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  5. 2016Adversarial examples in the physical worldcited by 1,762
  6. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,324
  7. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
  8. 2018Robustness May Be at Odds with Accuracycited by 873
  9. 2019Square Attack: a query-efficient black-box adversarial attack via random searchcited by 682
  10. 2017ZOO: Zeroth Order Optimization based Black-box Attacks to Deep Neural Networks without Training Substitute Modelscited by 656
  11. 2017Provable defenses against adversarial examples via the convex outer adversarial polytopecited by 564
  12. 2020On Adaptive Attacks to Adversarial Example Defensescited by 547
16 more
  1. 2018Black-box Adversarial Attacks with Limited Queries and Informationcited by 499
  2. 2018Certified Robustness to Adversarial Examples with Differential Privacycited by 462
  3. 2016Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samplescited by 439
  4. 2016Direct Feedback Alignment Provides Learning in Deep Neural Networkscited by 240
  5. 2018Prior Convictions: Black-Box Adversarial Attacks with Bandits and Priorscited by 145
  6. 2018Scaling provable adversarial defensescited by 141
  7. 2020Direct Feedback Alignment Scales to Modern Deep Learning Tasks and Architecturescited by 56
  8. 2019Parsimonious Black-Box Adversarial Attacks via Efficient Combinatorial Optimizationcited by 49
  9. 2020Adversarial Attacks on Linear Contextual Banditscited by 34
  10. 2019Theoretical evidence for adversarial robustness through randomizationcited by 33
  11. 2019Principled Training of Neural Networks with Direct Feedback Alignmentcited by 23
  12. 2019Yet another but more efficient black-box adversarial attack: tiling and evolution strategiescited by 17
  13. 2020Advocating for Multiple Defense Strategies against Adversarial Examplescited by 11
  14. 2019On the Adversarial Robustness of Neural Networks without Weight Transportcited by 5
  15. 2019Don't take it lightly: Phasing optical random projections with unknown operatorscited by 4
  16. 2019Fast Optical System Identification by Numerical Interferometrycited by 4

Built on by

Cited by year

Citing papers per year: 2021: 3; 2022: 2; 2023: 1; 2024: 0; 2025: 1.

7 papers.

  1. 2023Nonlinear optical encoding enabled by recurrent linear scatteringcited by 14
  2. 2021Benchmarking the Accuracy and Robustness of Feedback Alignment Algorithmscited by 9
  3. 2022Scaling Laws Beyond Backpropagationcited by 4
  4. 2021ROPUST: Improving Robustness through Fine-tuning with Photonic Processors and Synthetic Gradientscited by 2
  5. 2022Walking Noise: On Layer-Specific Robustness of Neural Architectures against Noisy Computations and Associated Characteristic Learning Dynamicscited by 2
  6. 2025Variance-Aware Noisy Training: Hardening DNNs against Unstable Analog Computationscited by 2
  7. 2021On the Noise Stability and Robustness of Adversarially Trained Networks on NVM Crossbarscited by 0