Is Your Data Relevant?: Dynamic Selection of Relevant Data for Federated Learning

Abstract

Federated Learning (FL) is a machine learning paradigm in which multiple clients participate to collectively learn a global machine learning model at the central server. It is plausible that not all the data owned by each client is relevant to the server's learning objective. The updates incorporated from irrelevant data could be detrimental to the global model. The task of selecting relevant data is explored in traditional machine learning settings where the assumption is that all the data is available in one place. In FL settings, the data is distributed across multiple clients and the server can't introspect it. This precludes the application of traditional solutions to selecting relevant data here. In this paper, we propose an approach called Federated Learning with Relevant Data (FLRD), that facilitates clients to derive updates using relevant data. Each client learns a model called Relevant Data Selector (RDS) that is private to itself to do the selection. This in turn helps in building an effective global model. We perform experiments with multiple real-world datasets to demonstrate the efficacy of our solution. The results show (a) the capability of FLRD to identify relevant data samples at each client locally and (b) the superiority of the global model learned by FLRD over other baseline algorithms.

Empowering Patient Similarity Networks through Innovative Data-Quality-Aware Federated Profiling

“…Traditional machine learning settings explore the subject of relevant data selection under the presumption that all relevant data are available in one location for computation. Due to the fact that the data in FL settings are dispersed among numerous clients and the server is unable to inspect it because of privacy restrictions [ 28 ], the conventional methods for choosing pertinent data are not applicable in this situation.…”

“…The data added to the ledger are verified for accuracy by the proof of common interest. In [ 28 ], it is suggested to use a method called federated learning with relevant data (FLRD), which allows clients to select data relevant to the server’s objective, leading to improved performance of the global learned model. Each client must learn a function that estimates data point relevance without compromising privacy.…”