Optimizing Federated Learning in Distributed Industrial IoT: A Multi-Agent Approach

In this paper, we aim to make the best joint decision of device selection and computing and spectrum resource allocation for optimizing federated learning (FL) performance in distributed industrial Internet of Things (IIoT) networks. To implement efficient FL over geographically dispersed data, we introduce a three-layer collaborative FL architecture to support deep neural network (DNN) training. Specifically, using the data dispersed in IIoT devices, the industrial gateways locally train the DNN model and the local models can be aggregated by their associated edge servers every FL epoch or by a cloud server every a few FL epochs for obtaining the global model. To optimally select participating devices and allocate computing and spectrum resources for training and transmitting the model parameters, we formulate a stochastic optimization problem with the objective of minimizing FL evaluating loss while satisfying delay and long-term energy consumption requirements. Since the objective function of the FL evaluating loss is implicit and the energy consumption is temporally correlated, it is difficult to solve the problem via traditional optimization methods. Thus, we propose a “Reinforcement on Federated” (RoF) scheme, based on deep multi-agent reinforcement learning, to solve the problem. Specifically, the RoF scheme is executed decentralizedly at edge servers, which can cooperatively make the optimal device selection and resource allocation decisions. Moreover, a device refinement subroutine is embedded into the RoF scheme to accelerate convergence while effectively saving the on-device energy. Simulation results demonstrate that the RoF scheme can facilitate efficient FL and achieve better performance compared with state-of-the-art benchmarks.

I. Introduction

With the wide deployment of industrial Internet of Things (IIoT), numerous devices are connected to the Internet, and voluminous industrial data is generated at the network edge [2]. As reported by Cisco, 2.4 TB data is burst out per minute from an industrial company in 2021 [3]. Utilized tremendous data, deep learning approaches, especially deep neural networks (DNNs) can facilitate industrial intelligence [4]. For example, a DNN model can be trained to provide fault diagnostic services for industrial facilities. In industrial scenarios, different factories usually generate a huge amount of data that contains critical information, such as the operation conditions of industrial facilities. Since this data information generally plays a crucial role in safety pre-warning and fault diagnostic, it is of great significance to preserve the privacy of industrial data. Traditional centralized training method needs to collect massive raw data from IIoT devices, e.g., industrial gateways (IGWs), which leads to user privacy concerns [5]. Hence, taking into account devices’ privacy-preserving requirements, the data would be better processed locally [6]–[8].