ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Towards Achieving Adversarial Robustness by Enforcing Feature Consistency across Bit Planes

Addepalli, S and Vivek, BS and Baburaj, A and Sriramanan, G and Venkatesh Babu, R (2020) Towards Achieving Adversarial Robustness by Enforcing Feature Consistency across Bit Planes. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 14-19 JUne 2020, Virtual, Online, pp. 1017-1026.

[img]
Preview
PDF
CVPR_2020.pdf - Published Version

Download (866kB) | Preview
Official URL: https://doi.org/10.1109/CVPR42600.2020.00110

Abstract

As humans, we inherently perceive images based on their predominant features, and ignore noise embedded within lower bit planes. On the contrary, Deep Neural Networks are known to confidently misclassify images corrupted with meticulously crafted perturbations that are nearly imperceptible to the human eye. In this work, we attempt to address this problem by training networks to form coarse impressions based on the information in higher bit planes, and use the lower bit planes only to refine their prediction. We demonstrate that, by imposing consistency on the representations learned across differently quantized images, the adversarial robustness of networks improves significantly when compared to a normally trained model. Present state-of-The-Art defenses against adversarial attacks require the networks to be explicitly trained using adversarial samples that are computationally expensive to generate. While such methods that use adversarial training continue to achieve the best results, this work paves the way towards achieving robustness without having to explicitly train on adversarial samples. The proposed approach is therefore faster, and also closer to the natural learning process in humans. © 2020 IEEE.

Item Type: Conference Paper
Publication: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Publisher: IEEE Computer Society
Additional Information: The copyright for this article belongs to the Authors.
Keywords: Deep neural networks; Pattern recognition, Bit planes; Feature consistency; Human eye; Learning process; State of the art; Training network, Image enhancement
Department/Centre: Division of Interdisciplinary Sciences > Computational and Data Sciences
Date Deposited: 24 Jan 2023 04:29
Last Modified: 24 Jan 2023 04:29
URI: https://eprints.iisc.ac.in/id/eprint/79294

Actions (login required)

View Item View Item