Key protected classification for collaborative learning

Creative Commons License

Sariyildiz M. B., Cinbi R. G., Ayday E.

Pattern Recognition, vol.104, 2020 (SCI-Expanded) identifier identifier identifier

  • Publication Type: Article / Article
  • Volume: 104
  • Publication Date: 2020
  • Doi Number: 10.1016/j.patcog.2020.107327
  • Journal Name: Pattern Recognition
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, PASCAL, Applied Science & Technology Source, BIOSIS, Compendex, Computer & Applied Sciences, INSPEC, MLA - Modern Language Association Database, zbMATH
  • Keywords: Privacy-preserving machine learning, collaborative learning, classification, generative adversarial networks
  • Middle East Technical University Affiliated: Yes

Abstract

© 2020Large-scale datasets play a fundamental role in training deep learning models. However, dataset collection is difficult in domains that involve sensitive information. Collaborative learning techniques provide a privacy-preserving solution, by enabling training over a number of private datasets that are not shared by their owners. However, recently, it has been shown that the existing collaborative learning frameworks are vulnerable to an active adversary that runs a generative adversarial network (GAN) attack. In this work, we propose a novel classification model that is resilient against such attacks by design. More specifically, we introduce a key-based classification model and a principled training scheme that protects class scores by using class-specific private keys, which effectively hide the information necessary for a GAN attack. We additionally show how to utilize high dimensional keys to improve the robustness against attacks without increasing the model complexity. Our detailed experiments demonstrate the effectiveness of the proposed technique. Source code will be made available at https://github.com/mbsariyildiz/key-protected-classification.