Privacy-preserving Decentralized Aggregation for Federated Learning

Federated learning is a promising framework for learning over decentralized data spanning multiple regions. This approach avoids expensive central training data aggregation cost and can improve privacy because distributed sites do not have to reveal privacy-sensitive data. In this paper, we develop a privacy-preserving decentralized aggregation protocol for federated learning. We formulate the distributed aggregation protocol with the Alternating Direction Method of Multiplier (ADMM) and examine its privacy weakness. Unlike prior work that use Differential Privacy or homomorphic encryption for privacy, we develop a protocol that controls communication among participants in each round of aggregation to minimize privacy leakage. We establish its privacy guarantee against an honest-but-curious adversary. We also propose an efficient algorithm to construct such a communication pattern, inspired by combinatorial block design theory. Our secure aggregation protocol based on this novel group communication pattern design leads to an efficient algorithm for federated training with privacy guarantees. We evaluate our federated training algorithm on image classification and next-word prediction applications over benchmark datasets with 9 and 15 distributed sites. Evaluation results show that our algorithm performs comparably to the standard centralized federated learning method while preserving privacy; the degradation in test accuracy is only up to 0.73%.

Privacy-preserving Decentralized Federated Learning

In this paper, we develop SecureD-FL, a privacy-preserving decentralized federated learning algorithm, i.e., without the traditional centralized aggregation server. For the decentralized aggregation, we employ the Alternating Direction Method of Multiplier (ADMM) and examine its privacy weakness. To address the privacy risk, we introduce a communication pattern inspired by the combinatorial block design theory and establish its theoretical privacy guarantee. We also propose an efficient algorithm to construct such a communication pattern. We evaluate our method on image classification and next-word prediction applications over federated benchmark datasets with nine and fifteen distributed sites hosting training data. While preserving privacy, SecureD-FL performs comparably to the standard centralized federated learning method; the degradation in test accuracy is only up to 0.73%.