Towards federated learning: An overview of methods and applications

Abstract

Federated learning (FL) is a collaborative, decentralized privacy-preserving method to attach the challenges of storing data and data privacy. Artificial intelligence, machine learning, smart devices, and deep learning have strongly marked the last years. Two challenges arose in data science as a result. First, the regulation protected the data by creating the General Data Protection Regulation, in which organizations are not allowed to keep or transfer data without the owner's authorization. Another challenge is the large volume of data generated in the era of big data, and keeping that data in one only server becomes increasingly tricky. Therefore, the data is allocated into different locations or generated by devices, creating the need to build models or perform calculations without transferring data to a single location. The new term FL emerged as a sub-area of machine learning that aims to solve the challenge of making distributed models with privacy considerations. This survey starts by describing relevant concepts, definitions, and methods, followed by an indepth investigation of federated model evaluation. Finally, we discuss three promising applications for further research: anomaly detection, distributed data streams, and graph representation.

U.S.-U.K. PETs Prize Challenge: Anomaly Detection via Privacy-Enhanced Federated Learning

“…Privacy-preserving federated anomaly detection. Federated learning-based anomaly detection, allowing multiple clients to collaboratively train a global model without pooling local datasets, is a promising solution to improve outlier detection performance in various applications [42], [43]. Despite its advantages, FL systems were susceptible to several attacks and could compromise the privacy of local datasets [44].…”

MoFLeuR: Motion-based Federated Learning Gesture Recognition

“…However, transferring the local datasets to an edge server has high transmission costs and violates the client's privacy. Federated Learning (FL) is a new concept in ML that is used to train global models without sharing the local data [6], [7]. Instead of gathering the datasets in one silo, FL aggregates the trained local models of clients to construct a global model [8].…”