Research of Black-Box Adversarial Attack Detection Based on GAN
- Resource Type
- Conference
- Authors
- Gong, Jinhong; Ling, Shiyong
- Source
- 2023 2nd International Conference on Artificial Intelligence and Computer Information Technology (AICIT) Artificial Intelligence and Computer Information Technology (AICIT), 2023 2nd International Conference on. :1-5 Sep, 2023
- Subject
- Communication, Networking and Broadcast Technologies
Computing and Processing
Power, Energy and Industry Applications
Robotics and Control Systems
Signal Processing and Analysis
Training
Closed box
Training data
Machine learning
Benchmark testing
Generative adversarial networks
Malware
Black-box attack
Generate adversarial network
Abnormal behavior detection
- Language
The abnormal behavior detector based on machine learning is vulnerable to adversarial attacks. Attackers achieve adversarial attacks with constructing counter samples on the detection algorithm, and get the expected output for the target model, by changing the attack behavior pattern and API call sequence. In order to improve the performance of black-box attack, this paper can generate adversarial sample to deceive discriminator by training the generator and discriminator through the black-box adversarial attack network based on GAN. Experiments on real world API attacks and benchmark data sets have proved that the anti-attack anomaly behavior detection system can well detect malware and anomaly behavior.