Efficient and Privacy-Enhanced Federated Learning for Industrial Artificial Intelligence

By leveraging deep learning-based technologies, industrial artificial intelligence (IAI) has been applied to solve various industrial challenging problems in Industry 4.0. However, for privacy reasons, traditional centralized training may be unsuitable for sensitive data-driven industrial scenarios, such as healthcare and autopilot. Recently, federated learning has received widespread attention, since it enables participants to collaboratively learn a shared model without revealing their local data. However, studies have shown that, by exploiting the shared parameters adversaries can still compromise industrial applications such as auto-driving navigation systems, medical data in wearable devices, and industrial robots' decision making. In this article, to solve this problem, we propose an efficient and privacy-enhanced federated learning (PEFL) scheme for IAI. Compared with existing solutions, PEFL is noninteractive, and can prevent private data from being leaked even if multiple entities collude with each other. Moreover, extensive experiments with real-world data demonstrate the superiority of PEFL in terms of accuracy and efficiency.

I. Introduction

Industry 4.0 enhances the operational efficiency of the entire manufacturing process by incorporating multiple emerging technologies, including blockchain, Internet of Things, cloud computing, and artificial intelligence [1], [2]. Recently, industrial artificial intelligence (IAI) has attracted widespread attention and has driven the arrival of Industry 4.0 [3], [4]. As a branch of artificial intelligence, deep learning (DL) is widely used to solve data-driven industrial problems in real-world applications, such as smart grid, autonomous driving, and facial recognition [5], [6]. The development of DL-based products often requires large amount of data collected from different IoT devices, which is essential for learning high-quality DL models. Nevertheless, the collection of massive data for centralized training causes serious privacy threats [7], [8]. For example, malicious adversaries can compromise the privacy of IoT devices by eavesdropping industry data uploaded to cloud service (CS) providers (such as Microsoft Azure Machine Learning,¹

[Online]. Available: https://azure.microsoft.com.

Google Cloud ML Engine²

[Online]. Available: https://cloud.google.com/ml-engine/.

) [9], [10]. On the other hand, the large scale of current industrial systems, represented by the number of industrial nodes and the amount of data generated, makes such traditional deployment (e.g., centralized training) nontractable [11], [12]. Therefore, it puts new demands on how to effectively and collaboratively implement artificial intelligence in large-scale industrial applications.