Cohort-based Federated Learning Services for Industrial Collaboration on the Edge

Abstract

<div>Machine Learning (ML) is increasingly applied in industrial manufacturing, but often performance is limited due to insufficient training data. While ML models can benefit from collaboration, due to privacy concerns, individual manufacturers cannot share data directly. Federated Learning (FL) enables collaborative training of ML models without revealing raw data. However, current FL approaches fail to take the characteristics and requirements of industrial clients into account. In this work, we propose a FL system consisting of a process description and a software architecture to provide \acrfull{flaas} to industrial clients deployed to edge devices. Our approach deals with skewed data by organizing clients into cohorts with similar data distributions. We evaluated the system on two industrial datasets. We show how the FLaaS approach provides FL to client processes by considering their requests submitted to the Industrial Federated Learning (IFL) Services API. Experiments on both industrial datasets and different FL algorithms show that the proposed cohort building can increase the ML model performance notably.</div>

Litigating the Algorithmic Boss in the EU: A (Legally) Feasible and (Strategically) Attractive Option for Trade Unions?

“…Although the Courts in the EU are still classifying platform workers in different ways, there is now a prevailing trend in considering them as employees, especially in ride-hailing and food delivery services, where litigation efforts have been stronger. 184 However, within the platform economy, resorting to litigation has been part of a broader strategy, aimed at fulfilling meta-legal purposes, such as: mobilization of trade unionists; galvanizing others to join the cause, especially in scarcely unionized sectors such as those in which platforms operate; campaigning to raise social awareness and encouraging public debate on the risks connected to the increasing use of algorithmic management in platform work; and finally lobbying to influence lawmakers to adopt policies aiming at mitigating them. 185 In this respect, litigation has been used as a successful tool within a broader trade union strategy of industrial and political activism that has been a driver of social and legal change, both at national and EU level.…”

Data Heterogeneity-Robust Federated Learning via Group Client Selection in Industrial IoT

“…Few literature has been proposed to tackle the non-i.i.d. data challenge of FL in IIoT, such as approaches based on centroid distance weighted averaging [7], reinforcement learning [21], and kmeans-based cohorts [22]. However, none of them take into account the changing local data distribution or the natural geographic clustering property of devices in IIoT.…”

Explainable Federated Learning: A Lifecycle Dashboard for Industrial Settings

“…Afterward, the server applies an FL algorithm, for example, FedAvg [5], that averages all gathered model parameters and sends the new model back to the clients. The number of these exchanges, also known as communication rounds, can be predefined or dependent on the required model's performance [6]. FL has been used in the manufacturing industry to improve anomaly detection and condition monitoring, facilitated by various data sources, like edge devices equipped with Internet of Things (IoT) sensors [7].…”