Hardware-Assisted Intellectual Property Protection of Deep Learning Models
By Abhishek Chakraborty, Ankit Mondal. 2020
103 citations recorded by the source corpus.
The protection of intellectual property (IP) rights of well-trained deep learning (DL) models has become a matter of major concern, especially with the growing trend of deployment of Machine Learning as a Service (MLaaS). In this work, we demonstrate the utilization of a hardware root-of-trust to safeguard the IPs of such DL models which potential attackers have access to. We propose an obfuscation framework called Hardware Protected Neural Network (HPNN) in which a deep neural network is trained as a function of a secret key and then, the obfuscated DL model is hosted on a public model sharing platform. This framework ensures that only an authorized end-user who possesses a trustworthy hardware device (with the secret key embedded on-chip) is able to run intended DL applications using the published model. Extensive experimental evaluations show that any unauthorized usage of such obfuscated DL models result in significant accuracy drops ranging from 73.22 to 80.17% across different neural network architectures and benchmark datasets. In addition, we also demonstrate the robustness of proposed HPNN framework against a model fine-tuning type of attack.
Subfields: Privacy & Security
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Privacy & Security
- Irene Solaiman
- Xiao Wang
- Debdeep Mukhopadhyay
- Sayandeep Saha
- Lixu Wang
- Manaar Alam
- Qi Zhu
- Sandipan Kundu
Citations in the corpus
- The Gradient of Generative AI Release: Methods and Considerations (2023, background): “[23] proposed an obfuscation framework that only authorizes users with a trustworthy hardware device.”
- NN-Lock: A Lightweight Authorization to Prevent IP Threats of Deep Learning Models (2022, methodology): “It is reported in [Chakraborty et al. 2020] that with a similar-sized dataset, an adversary can ine-tune a DNN model locked with HPNN to achieve 82.”
- Non-Transferable Learning: A New Approach for Model Ownership Verification and Applicability Authorization (2021, methodology): “Chakraborty et al. (2020) generate a secret key from hardware fingerprints of a particular device, and require that only users who possess this device can load and employ the model.”