Attri-Fed: A GIB Framework for Attribute-Based Privacy and Communication-Efficient Federated Learning
- Resource Type
- Conference
- Authors
- Saz, Ahmet Faruk; Malur Saidutta, Yashas; Fekri, Faramarz; Akdeniz, Mustafa Riza; Edwards, Brandon; Himayat, Nageen
- Source
- 2023 IEEE 24th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC) Signal Processing Advances in Wireless Communications (SPAWC), 2023 IEEE 24th International Workshop on. :366-370 Sep, 2023
- Subject
- Communication, Networking and Broadcast Technologies
Signal Processing and Analysis
Wireless communication
Privacy
Differential privacy
Federated learning
Conferences
Signal processing
Servers
information bottleneck
differential privacy
federated learning
minimax optimization
variational bounds
- Language
- ISSN
- 1948-3252
We present an attribute-based privacy framework for Federated Learning. Our framework utilizes the Generalized Information Bottleneck (GIB) principle to create functionally compressed representations that obscure designated sensitive attributes from potential inferential adversaries at the server. These functional representations are generated following a minimax adversarial optimization of the privacy and utility bounds on the optimization function. We show that our proposed framework inherently provides attribute-based differential privacy (DP) guarantees, reduces communication overhead and improves utility (e.g., image classification performance). We presented both theoretical and experimental comparisons of our less restrictive attribute-based DP approach with conventional DP.