CoreLocker: Neuron-level Usage Control
Zihan Wang, Zhongkui Ma, Xinguo Feng, Ruoxi Sun, Hu Wang, Minhui Xue
IEEE Symposium on Security and Privacy 2024 · Day 2 · Continental Ballroom 5
The talk "CoreLocker: Neuron-level Usage Control" introduces a novel approach to protect the **intellectual property (IP)** of **deep neural networks (DNNs)** and enable their controlled monetization. Presented by Zihan Wang, a PhD student at the University of Queensland, the research addresses the escalating challenge of safeguarding valuable AI models against unauthorized usage, particularly when deployed in untrusted environments such as user devices or through **machine learning as a service (MLaaS)** platforms. The core problem lies in the significant investment required to develop powerful DNNs—exemplified by models like GPT-3, which demanded 355 GPU years and an estimated $4.6 million for a single training run—and the subsequent financial losses incurred when these models are exploited.
AI review
CoreLocker presents a novel neuron-level usage control for DNN IP protection, leveraging significant weights as an access key. It enables granular utility degradation and restoration for pre-trained models without retraining, offering a lightweight and versatile solution for secure monetization. Backed by solid theoretical bounds and empirical validation, it significantly advances model protection.