Concurrent vertical and horizontal federated learning with fuzzy cognitive maps

TL;DR

This paper proposes a novel federated learning framework using fuzzy cognitive maps to effectively handle diverse, non-IID data distributions across participants while preserving data privacy.

Contribution

It introduces a new federated learning approach employing fuzzy cognitive maps to address challenges of data heterogeneity and privacy in distributed settings.

Findings

Effective handling of non-IID data distributions.

Improved learning outcomes while maintaining privacy.

Versatile federation strategies tested successfully.

Abstract

Data privacy is a major concern in industries such as healthcare or finance. The requirement to safeguard privacy is essential to prevent data breaches and misuse, which can have severe consequences for individuals and organisations. Federated learning is a distributed machine learning approach where multiple participants collaboratively train a model without compromising the privacy of their data. However, a significant challenge arises from the differences in feature spaces among participants, known as non-IID data. This research introduces a novel federated learning framework employing fuzzy cognitive maps, designed to comprehensively address the challenges posed by diverse data distributions and non-identically distributed features in federated settings. The proposal is tested through several experiments using four distinct federation strategies: constant-based, accuracy-based, AUC-based, and precision-based weights. The results demonstrate the effectiveness of the approach in achieving the desired learning outcomes while maintaining privacy and confidentiality standards.