PRIVACY-ENHANCED ZERO-SHOT LEARNING VIA DATA-FREE KNOWLEDGE TRANSFER

Abstract

Considering the increasing concerns about data copyright and sensitivity issues, we present a novel Privacy-Enhanced Zero-Shot Learning (PE-ZSL) paradigm. The key innovation is to involve a teacher model as the data safeguard to guide the PE-ZSL model training without data sharing. The PE-ZSL model consists of a generator and student network, which can achieve data-free knowledge transfer while maintaining the performance of teacher model. We investigate ‘black-’ and ‘white-box’ scenarios in PE-ZSL task as different levels of framework privacy. Besides, we provide the discussion of teacher model in both omniscient and quasi-omniscient settings according to the knowledge space. Despite simple implementations and data-missing disadvantages, our PE-ZSL framework can retain state-of-the-art ZSL and GZSL performance under the ‘white-box’ scenario. Extensive qualitative and quantitative analysis also demonstrates promising results when deploying the model under ‘black-box’ scenario.

@INPROCEEDINGS{10219812,
  author={Gao, Rui and Wan, Fan and Organisciak, Daniel and Pu, Jiyao and Duan, Haoran and Zhang, Peng and Hou, Xingsong and Long, Yang},
  booktitle={2023 IEEE International Conference on Multimedia and Expo (ICME)}, 
  title={Privacy-Enhanced Zero-Shot Learning via Data-Free Knowledge Transfer}, 
  year={2023},
  volume={},
  number={},
  pages={432-437},
  keywords={Training;Privacy;Technological innovation;Sensitivity;Statistical analysis;Data models;Generators;Zero-Shot Learning;Privacy Protection;Data-Free Knowledge Transfer},
  doi={10.1109/ICME55011.2023.00081}}