HLF-FSL. A Decentralized Federated Split Learning Solution for IoT on Hyperledger Fabric
This addresses privacy and scalability issues for enterprise IoT deployments, though it is incremental as it builds on existing federated and split learning techniques.
The paper tackles the need for scalable, privacy-preserving collaborative machine learning in sensitive domains by proposing a decentralized architecture combining Federated Split Learning with Hyperledger Fabric, achieving matching accuracy to centralized methods while reducing training time on benchmarks like CIFAR-10 and MNIST.
Collaborative machine learning in sensitive domains demands scalable, privacy preserving solutions for enterprise deployment. Conventional Federated Learning (FL) relies on a central server, introducing single points of failure and privacy risks, while Split Learning (SL) partitions models for privacy but scales poorly due to sequential training. We present a decentralized architecture that combines Federated Split Learning (FSL) with the permissioned blockchain Hyperledger Fabric (HLF). Our chaincode orchestrates FSL's split model execution and peer-to-peer aggregation without any central coordinator, leveraging HLF's transient fields and Private Data Collections (PDCs) to keep raw data and model activations private. On CIFAR-10 and MNIST benchmarks, HLF-FSL matches centralized FSL accuracy while reducing per epoch training time compared to Ethereum-based works. Performance and scalability tests show minimal blockchain overhead and preserved accuracy, demonstrating enterprise grade viability.