Amicable Aid: Perturbing Images to Improve Classification Performance
Not on the map. Why?
Abstract (arXiv)
While adversarial perturbation of images to attack deep image classification models pose serious security concerns in practice, this paper suggests a novel paradigm where the concept of image perturbation can benefit classification performance, which we call amicable aid. We show that by taking the opposite search direction of perturbation, an image can be modified to yield higher classification confidence and even a misclassified image can be made correctly classified. This can be also achieved with a large amount of perturbation by which the image is made unrecognizable by human eyes. The mechanism of the amicable aid is explained in the viewpoint of the underlying natural image manifold. Furthermore, we investigate the universal amicable aid, i.e., a fixed perturbation can be applied to multiple images to improve their classification results. While it is challenging to find such perturbations, we show that making the decision boundary as perpendicular to the image manifold as possible via training with modified data is effective to obtain a model for which universal amicable perturbations are more easily found.
Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 30 of its 36 references that are in the map. First posted on arXiv 9 Dec 2021.
Builds on
30 of 36 references.
References · 30
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
- 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
- 2014Explaining and Harnessing Adversarial Examplescited by 7,287
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
- 2013Intriguing properties of neural networkscited by 5,114
- 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
- 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
- 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
- 2016Adversarial Machine Learning at Scalecited by 934
- 2017Ensemble Adversarial Training: Attacks and Defensescited by 833
- 2014Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Imagescited by 774
18 more
- 2016Universal adversarial perturbationscited by 774
- 2020Fast is better than free: Revisiting adversarial trainingcited by 723
- 2019Adversarial Examples Improve Image Recognitioncited by 283
- 2021Unlearnable Examples: Making Personal Data Unexploitablecited by 188
- 2018Disentangling Adversarial Robustness and Generalizationcited by 171
- 2018Constructing Unrestricted Adversarial Examples with Generative Modelscited by 135
- 2018Is Robustness the Cost of Accuracy? -- A Comprehensive Study on the Robustness of 18 Deep Image Classification Modelscited by 135
- 2018SparseFool: a few pixels make a big differencecited by 94
- 2019Deep Neural Network Fingerprinting by Conferrable Adversarial Examplescited by 90
- 2019ColorFool: Semantic Adversarial Colorizationcited by 89
- 2020On Success and Simplicity: A Second Look at Transferable Targeted Attackscited by 89
- 2019Towards Large yet Imperceptible Adversarial Image Perturbations with Perceptual Color Distancecited by 72
- 2016A Boundary Tilting Persepective on the Phenomenon of Adversarial Examplescited by 70
- 2020Perturbing Across the Feature Hierarchy to Improve Standard and Strict Blackbox Attack Transferabilitycited by 57
- 2021Targeted Attack against Deep Neural Networks via Flipping Limited Weight Bitscited by 49
- 2020Unadversarial Examples: Designing Objects for Robust Visioncited by 45
- 2018Towards Imperceptible and Robust Adversarial Example Attacks against Neural Networkscited by 38
- 2020Just One Moment: Structural Vulnerability of Deep Action Recognition against One Frame Attackcited by 12
Built on by
Citing papers per year: 2023: 1; 2024: 1; 2025: 0; 2026: 1.
3 papers.