FaceLeaks: Inference Attacks against Transfer Learning Models via Black-box Queries
arXiv:2010.14023
Abstract
Transfer learning is a useful machine learning framework that allows one to build task-specific models (student models) without significantly incurring training costs using a single powerful model (teacher model) pre-trained with a large amount of data. The teacher model may contain private data, or interact with private inputs. We investigate if one can leak or infer such private information without interacting with the teacher model directly. We describe such inference attacks in the context of face recognition, an application of transfer learning that is highly sensitive to personal privacy. Under black-box and realistic settings, we show that existing inference techniques are ineffective, as interacting with individual training instances through the student models does not reveal information about the teacher. We then propose novel strategies to infer from aggregate-level information. Consequently, membership inference attacks on the teacher model are shown to be possible, even when the adversary has access only to the student models. We further demonstrate that sensitive attributes can be inferred, even in the case where the adversary has limited auxiliary information. Finally, defensive strategies are discussed and evaluated. Our extensive study indicates that information leakage is a real privacy threat to the transfer learning framework widely used in real-life situations.
References in corpus (15)
- Deep Learning with Differential Privacy
- Deep Learning Face Representation by Joint Identification-Verification
- Comprehensive Privacy Analysis of Deep Learning: Passive and Active White-box Inference Attacks against Centralized and Federated Learning
- Stealing Machine Learning Models via Prediction APIs
- The Secret Sharer: Evaluating and Testing Unintended Memorization in Neural Networks
- Robust Physical-World Attacks on Deep Learning Models
- CosFace: Large Margin Cosine Loss for Deep Face Recognition
- Robust Audio Adversarial Example for a Physical Attack
- Stolen Memories: Leveraging Model Memorization for Calibrated White-Box Membership Inference
- Updates-Leak: Data Set Inference and Reconstruction Attacks in Online Learning
- Overlearning Reveals Sensitive Attributes
- Indirect Adversarial Attacks via Poisoning Neighbors for Graph Convolutional Networks
- Privacy Analysis of Deep Learning in the Wild: Membership Inference Attacks against Transfer Learning
- A Comprehensive Analysis of Information Leakage in Deep Transfer Learning
- P3GM: Private High-Dimensional Data Release via Privacy Preserving Phased Generative Model