Ruyi Ding (Northeastern University), Tong Zhou (Northeastern University), Lili Su (Northeastern University), Aidong Adam Ding (Northeastern University), Xiaolin Xu (Northeastern University), Yunsi Fei (Northeastern University)

Adapting pre-trained deep learning models to customized tasks has become a popular choice for developers to cope with limited computational resources and data volume. More specifically, probing--training a classifier on a pre-trained encoder--has been widely adopted in transfer learning, which helps to prevent overfitting and catastrophic forgetting. However, such generalizability of pre-trained encoders raises concerns about the potential misuse of probing for harmful applications, such as discriminatory speculation and warfare applications. In this work, we introduce EncoderLock, a novel applicability authorization method designed to protect pre-trained encoders from malicious probing, i.e., yielding poor performance on specified prohibited domains while maintaining their utility in authorized ones. Achieving this balance is challenging because of the opposite optimization objectives and the variety of downstream heads that adversaries can utilize adaptively. To address these challenges, EncoderLock employs two techniques: domain-aware weight selection and updating to restrict applications on prohibited domains/tasks, and self-challenging training scheme that iteratively strengthens resistance against any potential downstream classifiers that adversaries may apply. Moreover, recognizing the potential lack of data from prohibited domains in practical scenarios, we introduce three EncoderLock variants with different levels of data accessibility: supervised (prohibited domain data with labels), unsupervised (prohibited domain data without labels), and zero-shot (no data or labels available). Extensive experiments across fifteen domains and three model architectures demonstrate EncoderLock's effectiveness over baseline methods using non-transferable learning. Additionally, we verify EncoderLock's effectiveness and practicality with a real-world pre-trained Vision Transformer (ViT) encoder from Facebook. These results underscore the valuable contributions EncoderLock brings to the development of responsible AI.

View More Papers

Reinforcement Unlearning

Dayong Ye (University of Technology Sydney), Tianqing Zhu (City University of Macau), Congcong Zhu (City University of Macau), Derui Wang (CSIRO’s Data61), Kun Gao (University of Technology Sydney), Zewei Shi (CSIRO’s Data61), Sheng Shen (Torrens University Australia), Wanlei Zhou (City University of Macau), Minhui Xue (CSIRO's Data61)

Read More

Vulnerability, Where Art Thou? An Investigation of Vulnerability Management...

Daniel Klischies (Ruhr University Bochum), Philipp Mackensen (Ruhr University Bochum), Veelasha Moonsamy (Ruhr University Bochum)

Read More

“I’m 73, you can’t expect me to have multiple...

Ashley Sheil (Munster Technological University), Jacob Camilleri (Munster Technological University), Michelle O Keeffe (Munster Technological University), Melanie Gruben (Munster Technological University), Moya Cronin (Munster Technological University) and Hazel Murray (Munster Technological University)

Read More

Privacy-Preserving Data Deduplication for Enhancing Federated Learning of Language...

Aydin Abadi (Newcastle University), Vishnu Asutosh Dasu (Pennsylvania State University), Sumanta Sarkar (University of Warwick)

Read More