Incentivizing Truthful Collaboration in Heterogeneous Federated Learning

TL;DR

This paper addresses the challenge of incentivizing truthful gradient updates in heterogeneous federated learning by proposing a payment scheme that discourages manipulation, ensuring model integrity across various protocols and tasks.

Contribution

It introduces a game-theoretic payment rule that provably disincentivizes clients from manipulating updates in federated learning, with explicit bounds on payments and convergence.

Findings

The payment rule effectively prevents manipulation in multiple federated learning protocols.

Experimental results show reduced manipulation and maintained model performance.

The approach balances heterogeneity, payments, and convergence in federated learning.

Abstract

Federated learning (FL) is a distributed collaborative learning method, where multiple clients learn together by sharing gradient updates instead of raw data. However, it is well-known that FL is vulnerable to manipulated updates from clients. In this work we study the impact of data heterogeneity on clients' incentives to manipulate their updates. First, we present heterogeneous collaborative learning scenarios where a client can modify their updates to be better off, and show that these manipulations can lead to diminishing model performance. To prevent such modifications, we formulate a game in which clients may misreport their gradient updates in order to "steer" the server model to their advantage. We develop a payment rule that provably disincentivizes sending modified updates under the FedSGD protocol. We derive explicit bounds on the clients' payments and the convergence rate of the global model, which allows us to study the trade-off between heterogeneity, payments and convergence. Finally, we provide an experimental evaluation of the effectiveness of our payment rule in the FedSGD, median-based aggregation FedSGD and FedAvg protocols on three tasks in computer vision and natural language processing. In all cases we find that our scheme successfully disincentivizes modifications.