Incentivize Contribution and Learn Parameters Too: Federated Learning with Strategic Data Owners

TL;DR

This paper develops mechanisms for federated learning that incentivize data owners to contribute truthfully and fully, ensuring optimal model learning while considering their strategic behavior.

Contribution

It introduces two mechanisms that incentivize truthful contribution and full data sharing in federated learning, addressing strategic behavior and achieving near-optimal outcomes.

Findings

Mechanisms converge quickly on real datasets.

Achieve good welfare guarantees.

Improve model performance for all agents.

Abstract

Classical federated learning (FL) assumes that the clients have a limited amount of noisy data with which they voluntarily participate and contribute towards learning a global, more accurate model in a principled manner. The learning happens in a distributed fashion without sharing the data with the center. However, these methods do not consider the incentive of an agent for participating and contributing to the process, given that data collection and running a distributed algorithm is costly for the clients. The question of rationality of contribution has been asked recently in the literature and some results exist that consider this problem. This paper addresses the question of simultaneous parameter learning and incentivizing contribution in a truthful manner, which distinguishes it from the extant literature. Our first mechanism incentivizes each client to contribute to the FL process at a Nash equilibrium and simultaneously learn the model parameters. We also ensure that agents are incentivized to truthfully reveal information in the intermediate stages of the algorithm. However, this equilibrium outcome can be away from the optimal, where clients contribute with their full data and the algorithm learns the optimal parameters. We propose a second mechanism that enables the full data contribution along with optimal parameter learning. Large scale experiments with real (federated) datasets (CIFAR-10, FEMNIST, and Twitter) show that these algorithms converge quite fast in practice, yield good welfare guarantees and better model performance for all agents.