README.md

Adversarial Defense with Robust Principal Component Analysis and Wavelet Denoising

IEEE International Conference On Range Technology (ICORT) 2023

Aryaman Sinha, N. B. Puhan, et al.

Indian Institute of Technology Bhubaneswar

Link: https://ieeexplore.ieee.org/document/10249152

Abstract

In this work, a local robust principal component analysis (RPCA)-based adversarial defense methodology is presented. The proposed defense is inspired by recent works of input reconstruction paradigm. The proposed method uses localised processing to apply RPCA towards removal of adversarial perturbations. The reconstructed image is guided by applying wavelet denoising to remove residual perturbation which improves the defense more efficiently. Experimental results are presented using state-of-the-art C&W attack and Square attack to show the effectiveness of the proposed adversarial defense on a pre-trained classifier model.

Approach

Requirements

• Python 3.6
• TensorFlow 2.x
• MATLAB 2021a

Results

Contact

Do you have any problem or doubts please raise the issue or directly contact to Aryaman Sinha

Citation

@INPROCEEDINGS{10249152,
  author={Sinha, Aryaman and Puhan, Niladri B. and Dash, Soumya P. and Mrudhul, Guda and Panda, Ganapati},
  booktitle={2023 3rd International Conference on Range Technology (ICORT)}, 
  title={Adversarial Defense with Local Robust Principal Component Analysis and Wavelet Denoising}, 
  year={2023},
  volume={},
  number={},
  pages={1-6},
  doi={10.1109/ICORT56052.2023.10249152}}