Personalised Federated Learning On Heterogeneous Feature Spaces

TL;DR

This paper introduces FLIC, a federated learning framework that enables personalized models across clients with heterogeneous data representations by learning a common feature space through local embeddings and Wasserstein barycenters.

Contribution

The paper proposes a novel federated learning framework, FLIC, that handles heterogeneous feature spaces by learning a shared feature space with distribution alignment and Wasserstein barycenters.

Findings

FLIC outperforms existing benchmarks with heterogeneous features.

Theoretical analysis supports the effectiveness of the approach.

Framework successfully aligns diverse client data representations.

Abstract

Most personalised federated learning (FL) approaches assume that raw data of all clients are defined in a common subspace i.e. all clients store their data according to the same schema. For real-world applications, this assumption is restrictive as clients, having their own systems to collect and then store data, may use heterogeneous data representations. We aim at filling this gap. To this end, we propose a general framework coined FLIC that maps client's data onto a common feature space via local embedding functions. The common feature space is learnt in a federated manner using Wasserstein barycenters while the local embedding functions are trained on each client via distribution alignment. We integrate this distribution alignement mechanism into a federated learning approach and provide the algorithmics of FLIC. We compare its performances against FL benchmarks involving heterogeneous input features spaces. In addition, we provide theoretical insights supporting the relevance of our methodology.