Blind Federated Learning without initial model

TL;DR

This paper introduces a novel federated learning approach for Fuzzy Cognitive Maps that operates without an initial model, enhancing privacy and performance across multiple datasets.

Contribution

It proposes two Particle Swarm Optimization-based methods for blind federated learning of Fuzzy Cognitive Maps, eliminating the need for an initial model.

Findings

Improved accuracy and precision on open datasets

Effective privacy-preserving federated learning methodology

Demonstrated feasibility of blind federated learning

Abstract

Federated learning is an emerging machine learning approach that allows the construction of a model between several participants who hold their own private data. This method is secure and privacy-preserving, suitable for training a machine learning model using sensitive data from different sources, such as hospitals. In this paper, the authors propose two innovative methodologies for Particle Swarm Optimisation-based federated learning of Fuzzy Cognitive Maps in a privacy-preserving way. In addition, one relevant contribution this research includes is the lack of an initial model in the federated learning process, making it effectively blind. This proposal is tested with several open datasets, improving both accuracy and precision.