Hierarchically Fair Federated Learning

TL;DR

This paper introduces a hierarchically fair federated learning framework that incentivizes agents based on their contributions, ensuring fairness and encouraging participation among competitive, self-interested agents.

Contribution

The paper proposes a novel hierarchical fairness framework for federated learning, including an extension for heterogeneous models, with theoretical and empirical validation.

Findings

HFFL effectively promotes fairness among agents.

HFFL+ accommodates heterogeneous models.

Frameworks improve participation in competitive settings.

Abstract

When the federated learning is adopted among competitive agents with siloed datasets, agents are self-interested and participate only if they are fairly rewarded. To encourage the application of federated learning, this paper employs a management strategy, i.e., more contributions should lead to more rewards. We propose a novel hierarchically fair federated learning (HFFL) framework. Under this framework, agents are rewarded in proportion to their pre-negotiated contribution levels. HFFL+ extends this to incorporate heterogeneous models. Theoretical analysis and empirical evaluation on several datasets confirm the efficacy of our frameworks in upholding fairness and thus facilitating federated learning in the competitive settings.