Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
The paper introduces MOSAIC-FL, a micro-service based framework for Federated Learning (FL) that prioritizes privacy and security in sensitive domains like genomics. It employs a modular architecture with an efficient gRPC communication layer and a fault-tolerant secure aggregation protocol based on a Threshold variant of the CKKS homomorphic cryptosystem, enabling robust model aggregation while minimizing communication overhead. The framework's effectiveness is validated through various applications, including genomic classification tasks, demonstrating its capability to handle privacy-preserving data analysis in real-world scenarios.
A micro-service architecture that ensures privacy in Federated Learning can effectively tackle sensitive genomic data challenges without compromising performance.
Security and privacy are primordial requirements for Federated Learning (FL), especially in fields such as healthcare and genomics where sensitive information has to be analyzed. Our FL framework is designed to address these challenges while proposing a modular, flexible and micro-service architecture. More precisely, it integrates an efficient gRPC communication layer and a Finite State Machine to ensure robust component synchronization and threat detection, while relying on a fault-tolerant secure aggregation protocol using a Threshold variant of the CKKS homomorphic cryptosystem. This allows blind model aggregation by an orchestration server, requiring a minimum of $t$-out-of-$N$ active clients for decryption while minimizing communication overhead thanks to both cryptographic and network protocols. We ensure IND-CPA-D security through noise flooding and mitigate the recent key-recovery attack on synchronized decryptors by renewing the collective key material at every round. We demonstrate the framework's effectiveness through diverse use cases, ranging from standard image recognition (EMNIST) to complex genomic classification including breast cancer subtyping on TCGA, evaluating system performance across different threshold values and model scales.