Improved Gradient-Based Adversarial Attacks for Quantized Networks

Authors

  • Kartik Gupta Australian National University, DATA61-CSIRO
  • Thalaiyasingam Ajanthan Australian National University, Amazon Science

DOI:

https://doi.org/10.1609/aaai.v36i6.20637

Keywords:

Machine Learning (ML), Computer Vision (CV)

Abstract

Neural network quantization has become increasingly popular due to efficient memory consumption and faster computation resulting from bitwise operations on the quantized networks. Even though they exhibit excellent generalization capabilities, their robustness properties are not well-understood. In this work, we systematically study the robustness of quantized networks against gradient based adversarial attacks and demonstrate that these quantized models suffer from gradient vanishing issues and show a fake sense of robustness. By attributing gradient vanishing to poor forward-backward signal propagation in the trained network, we introduce a simple temperature scaling approach to mitigate this issue while preserving the decision boundary. Despite being a simple modification to existing gradient based adversarial attacks, experiments on multiple image classification datasets with multiple network architectures demonstrate that our temperature scaled attacks obtain near-perfect success rate on quantized networks while outperforming original attacks on adversarially trained models as well as floating-point networks.

Downloads

Published

2022-06-28

How to Cite

Gupta, K., & Ajanthan, T. (2022). Improved Gradient-Based Adversarial Attacks for Quantized Networks. Proceedings of the AAAI Conference on Artificial Intelligence, 36(6), 6810-6818. https://doi.org/10.1609/aaai.v36i6.20637

Issue

Section

AAAI Technical Track on Machine Learning I