Asynchronous Federated Learning System Based on Permissioned Blockchains.

Rong Wang,Wei-Tek Tsai

doi:10.3390/s22041672

Abstract

The existing federated learning framework is based on the centralized model coordinator, which still faces serious security challenges such as device differentiated computing power, single point of failure, poor privacy, and lack of Byzantine fault tolerance. In this paper, we propose an asynchronous federated learning system based on permissioned blockchains, using permissioned blockchains as the federated learning server, which is composed of a main-blockchain and multiple sub-blockchains, with each sub-blockchain responsible for partial model parameter updates and the main-blockchain responsible for global model parameter updates. Based on this architecture, a federated learning asynchronous aggregation protocol based on permissioned blockchain is proposed that can effectively alleviate the synchronous federated learning algorithm by integrating the learned model into the blockchain and performing two-order aggregation calculations. Therefore, the overhead of synchronization problems and the reliability of shared data is also guaranteed. We conducted some simulation experiments and the experimental results showed that the proposed architecture could maintain good training performances when dealing with a small number of malicious nodes and differentiated data quality, which has good fault tolerance, and can be applied to edge computing scenarios.

Highlights

The machine learning method is based on sample data training to obtain machine learning models suitable for different tasks and scenarios
Liu Jianchun et al [23] presented a new communication-efficient asynchronous federated learning (CE-AFL) mechanism in which the parameter server will only aggregate local model updates from a certain fraction α (0 < α < 1) of all edge nodes in the order of their arrival in each epoch
The asynchronous federated learning system based on permissioned blockchains is divided into four layers of architecture: the IoT device layer, the network layer, the edge computing layer, the blockchain layer, and the application layer

Summary

Introduction

The machine learning method is based on sample data training to obtain machine learning models suitable for different tasks and scenarios. Through the distributed ledger feature of the permissioned blockchains, it is naturally guaranteed that the model parameter data consistency, synchronization, and sharing between multiple participants in the federation learning are secure and trustworthy, and that the model parameter data interaction is transparent, traceable, tamper-proof, and anti-forgery. To address these challenges, we propose an asynchronous federal learning system based on permissioned blockchains that addresses the federated learning single point of failure problem and data security privacy issues.

Federated Learning

Blockchain-Based Federated Learning

System Overview

Node Selection Algorithm

11: Update actor2 with the parameters in actor1

Experimental Results