Honggang Yu (University of Florida), Kaichen Yang (University of Florida), Teng Zhang (University of Central Florida), Yun-Yun Tsai (National Tsing Hua University), Tsung-Yi Ho (National Tsing Hua University), Yier Jin (University of Florida)

Cloud-based Machine Learning as a Service (MLaaS) is gradually gaining acceptance as a reliable solution to various real-life scenarios. These services typically utilize Deep Neural Networks (DNNs) to perform classification and detection tasks and are accessed through Application Programming Interfaces (APIs). Unfortunately, it is possible for an adversary to steal models from cloud-based platforms, even with black-box constraints, by repeatedly querying the public prediction API with malicious inputs. In this paper, we introduce an effective and efficient black-box attack methodology that extracts largescale DNN models from cloud-based platforms with near-perfect performance. In comparison to existing attack methods, we significantly reduce the number of queries required to steal the target model by incorporating several novel algorithms, including active learning, transfer learning, and adversarial attacks. During our experimental evaluations, we validate our proposed model for conducting theft attacks on various commercialized MLaaS platforms including two Microsoft Custom Vision APIs (Microsoft Traffic Recognition API and Microsoft Flower Recognition API), the Face++ Emotion Recognition API, the IBM Watson Visual Recognition API, Google AutoML API, and the Clarifai Safe for Work (NSFW) API. Our results demonstrate that the proposed method can easily reveal/steal large-scale DNN models from these cloud platforms. Further, the proposed attack method can also be used to accurately evaluates the robustness of DNN based MLaaS image classifiers against theft attacks.

View More Papers

Metal: A Metadata-Hiding File-Sharing System

Weikeng Chen (UC Berkeley), Raluca Ada Popa (UC Berkeley)

Read More

Precisely Characterizing Security Impact in a Flood of Patches...

Qiushi Wu (University of Minnesota), Yang He (University of Minnesota), Stephen McCamant (University of Minnesota), Kangjie Lu (University of Minnesota)

Read More

Finding Safety in Numbers with Secure Allegation Escrows

Venkat Arun (Massachusetts Institute of Technology), Aniket Kate (Purdue University), Deepak Garg (Max Planck Institute for Software Systems), Peter Druschel (Max Planck Institute for Software Systems), Bobby Bhattacharjee (University of Maryland)

Read More

Mind the Portability: A Warriors Guide through Realistic Profiled...

Shivam Bhasin (Nanyang Technological University), Anupam Chattopadhyay (Nanyang Technological University), Annelie Heuser (Univ Rennes, Inria, CNRS, IRISA), Dirmanto Jap (Nanyang Technological University), Stjepan Picek (Delft University of Technology), Ritu Ranjan Shrivastwa (Secure-IC)

Read More