ATHENA: A Framework based on Diverse Weak Defenses for Building Adversarial Defense

Authors: Ying Meng, Jianhai Su, Jason O’Kane, Pooyan Jamshidi
Published: 2020-01-02 | Updated: 2020-10-16

Source: https://arxiv.org/abs/2001.00308

PDF: https://arxiv.org/pdf/2001.00308

Labels Predicted by AI

敵対的学習 防御手法の効果分析 透かし評価

Please note that these labels were automatically added by AI. Therefore, they may not be entirely accurate.
For more details, please see the About the Literature Database page.

Abstract

There has been extensive research on developing defense techniques against adversarial attacks; however, they have been mainly designed for specific model families or application domains, therefore, they cannot be easily extended. Based on the design philosophy of ensemble of diverse weak defenses, we propose ATHENA—a flexible and extensible framework for building generic yet effective defenses against adversarial attacks. We have conducted a comprehensive empirical study to evaluate several realizations of ATHENA with four threat models including zero-knowledge, black-box, gray-box, and white-box. We also explain (i) why diversity matters, (ii) the generality of the defense framework, and (iii) the overhead costs incurred by ATHENA.