Timezone: »

 
Poster
CAFE: Catastrophic Data Leakage in Vertical Federated Learning
Xiao Jin · Pin-Yu Chen · Chia-Yi Hsu · Chia-Mu Yu · Tianyi Chen

Thu Dec 09 08:30 AM -- 10:00 AM (PST) @
in Poster Session 6 »

Recent studies show that private training data can be leaked through the gradients sharing mechanism deployed in distributed machine learning systems, such as federated learning (FL). Increasing batch size to complicate data recovery is often viewed as a promising defense strategy against data leakage. In this paper, we revisit this defense premise and propose an advanced data leakage attack with theoretical justification to efficiently recover batch data from the shared aggregated gradients. We name our proposed method as catastrophic data leakage in vertical federated learning (CAFE). Comparing to existing data leakage attacks, our extensive experimental results on vertical FL settings demonstrate the effectiveness of CAFE to perform large-batch data leakage attack with improved data recovery quality. We also propose a practical countermeasure to mitigate CAFE. Our results suggest that private data participated in standard FL, especially the vertical case, have a high risk of being leaked from the training gradients. Our analysis implies unprecedented and practical data leakage risks in those learning settings. The code of our work is available at https://github.com/DeRafael/CAFE.

Keywords: Machine Learning · Privacy · Federated Learning

Author Information

Xiao Jin (Rensselaer Polytechnic Institute)
Pin-Yu Chen (IBM Research AI)
Chia-Yi Hsu (National Chiao Tung University)
Chia-Mu Yu (National Yang Ming Chiao Tung University)
Tianyi Chen (Rensselaer Polytechnic Institute)

More from the Same Authors