Prior-Independent Auctions for the Demand Side of Federated Learning

TL;DR

This paper introduces FIPIA, a novel auction-based mechanism for incentivizing participation in federated learning without prior knowledge of clients' interests, ensuring fair resource allocation.

Contribution

The paper proposes FIPIA, a prior-independent auction mechanism for federated learning that handles heterogeneous client interests without requiring prior information.

Findings

FIPIA effectively incentivizes clients in federated learning.

Experimental results show FIPIA's robustness across datasets.

Clients' model quality improves with FIPIA's incentives.

Abstract

Federated learning (FL) is a paradigm that allows distributed clients to learn a shared machine learning model without sharing their sensitive training data. While largely decentralized, FL requires resources to fund a central orchestrator or to reimburse contributors of datasets to incentivize participation. Inspired by insights from prior-independent auction design, we propose a mechanism, FIPIA (Federated Incentive Payments via Prior-Independent Auctions), to collect monetary contributions from self-interested clients. The mechanism operates in the semi-honest trust model and works even if clients have a heterogeneous interest in receiving high-quality models, and the server does not know the clients' level of interest. We run experiments on the MNIST, FashionMNIST, and CIFAR-10 datasets to test clients' model quality under FIPIA and FIPIA's incentive properties.