Conference proceeding
Reinforced Adversarial Attacks on Deep Neural Networks Using ADMM
2018 IEEE GLOBAL CONFERENCE ON SIGNAL AND INFORMATION PROCESSING (GLOBALSIP 2018), pp 1169-1173
01 Jan 2018
Featured in Collection : UN Sustainable Development Goals @ Drexel
Abstract
As deep learning penetrates into wide application domains, it is essential to evaluate the robustness of deep neural networks (DNNs) under adversarial attacks, especially for some security-critical applications. To better understand the security properties of DNNs, we propose a general framework for constructing adversarial examples, based on ADMM (Alternating Direction Method of Multipliers). This general framework can be adapted to implement L2 and L0 attacks with minor changes. Our ADMM attacks require less distortion for incorrect classification compared with C&W attacks. Our ADMM attack is also able to break defenses such as defensive distillation and adversarial training, and provide strong attack transferability.
Metrics
Details
- Title
- Reinforced Adversarial Attacks on Deep Neural Networks Using ADMM
- Creators
- Pu Zhao - Northeastern UniversityKaidi Xu - Northeastern UniversityTianyun Zhang - Syracuse UniversityMakan Fardad - Syracuse UniversityYanzhi Wang - Northeastern UniversityXue Lin - Northeastern University
- Publication Details
- 2018 IEEE GLOBAL CONFERENCE ON SIGNAL AND INFORMATION PROCESSING (GLOBALSIP 2018), pp 1169-1173
- Series
- IEEE Global Conference on Signal and Information Processing
- Publisher
- IEEE
- Number of pages
- 5
- Grant note
- CAREER CMMI-1750531; ECCS-1609916 / National Science Foundation; National Science Foundation (NSF) U.S. Office of Naval Research; Office of Naval Research FA8750-18-2-0058 / Air Force Research Laboratory
- Resource Type
- Conference proceeding
- Language
- English
- Academic Unit
- Computer Science (Computing)
- Web of Science ID
- WOS:000462968100240
- Scopus ID
- 2-s2.0-85063101070
- Other Identifier
- 991021871466604721
UN Sustainable Development Goals (SDGs)
This publication has contributed to the advancement of the following goals:
InCites Highlights
Data related to this publication, from InCites Benchmarking & Analytics tool:
- Collaboration types
- Domestic collaboration
- Web of Science research areas
- Engineering, Electrical & Electronic