NASPY: automated extraction of automated machine learning models

• Main Authors: Lou, Xiaoxuan, Guo, Shangwei, Li, Jiwei, Wu, Yaoxin, Zhang, Tianwei
• Other Authors: School of Computer Science and Engineering
• Format: Conference or Workshop Item
• Language: English
• Published: 2023
• Subjects: Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence; Automated Machine Learning; Deep Neural Networks
• Online Access: https://hdl.handle.net/10356/165393; https://openreview.net/group?id=ICLR.cc/2022/Conference#spotlight-submissions
• Institution: Nanyang Technological University

We present NASPY, an end-to-end adversarial framework to extract the networkarchitecture of deep learning models from Neural Architecture Search (NAS). Existing works about model extraction attacks mainly focus on conventional DNN models with very simple operations, or require heavy manual analysis...