A Heterogeneous Federated Transfer Learning Approach with Extreme Aggregation and Speed

• Published in: Mathematics (Basel) Vol. 10; no. 19; p. 3528
• Main Authors: Berghout, Tarek, Bentrcia, Toufik, Ferrag, Mohamed Amine, Benbouzid, Mohamed
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.10.2022; MDPI
• Subjects: Algorithms; Communication; Data security; Datasets; Deep learning; Engineering Sciences; Experiments; Federated learning; Federated searching; federated transfer learning; Heterogeneity; heterogeneous systems; Machine learning; Methods; Microprocessors; non identical independent data; Performance evaluation; Privacy; recursive least squares; Training; Algeria; federated transfer learning; non identical independent data; recursive least squares; heterogeneous systems; federated learning
• ISSN: 2227-7390; 2227-7390
• DOI: 10.3390/math10193528

Federated learning (FL) is a data-privacy-preserving, decentralized process that allows local edge devices of smart infrastructures to train a collaborative model independently while keeping data localized. FL algorithms, encompassing a well-structured average of the training parameters (e.g., the weights and biases resulting from training-based stochastic gradient descent variants), are subject to many challenges, namely expensive communication, systems heterogeneity, statistical heterogeneity, and privacy concerns. In this context, our paper targets the four aforementioned challenges while focusing on reducing communication and computational costs by involving recursive least squares (RLS) training rules. Accordingly, to the best of our knowledge, this is the first time that the RLS algorithm is modified to completely accommodate non-independent and identically distributed data (non-IID) for federated transfer learning (FTL). Furthermore, this paper also introduces a newly generated dataset capable of emulating such real conditions and of making data investigation available on ordinary commercial computers with quad-core microprocessors and less need for higher computing hardware. Applications of FTL-RLS on the generated data under different levels of complexity closely related to different levels of cardinality lead to a variety of conclusions supporting its performance for future uses.