Strategic Incentivization for Locally Differentially Private Federated Learning

TL;DR

This paper introduces a game-theoretic incentivization mechanism in federated learning that balances privacy and accuracy by rewarding clients based on their gradient perturbation levels, aiming to optimize overall model performance.

Contribution

It models the privacy-accuracy trade-off as a strategic game and proposes a token-based incentive mechanism to encourage clients to add less noise, improving model accuracy.

Findings

The incentivization mechanism effectively balances privacy and accuracy.

Clients are motivated to reduce noise through token rewards.

Experimental results demonstrate improved model performance with the proposed approach.

Abstract

In Federated Learning (FL), multiple clients jointly train a machine learning model by sharing gradient information, instead of raw data, with a server over multiple rounds. To address the possibility of information leakage in spite of sharing only the gradients, Local Differential Privacy (LDP) is often used. In LDP, clients add a selective amount of noise to the gradients before sending the same to the server. Although such noise addition protects the privacy of clients, it leads to a degradation in global model accuracy. In this paper, we model this privacy-accuracy trade-off as a game, where the sever incentivizes the clients to add a lower degree of noise for achieving higher accuracy, while the clients attempt to preserve their privacy at the cost of a potential loss in accuracy. A token based incentivization mechanism is introduced in which the quantum of tokens credited to a client in an FL round is a function of the degree of perturbation of its gradients. The client can later access a newly updated global model only after acquiring enough tokens, which are to be deducted from its balance. We identify the players, their actions and payoff, and perform a strategic analysis of the game. Extensive experiments were carried out to study the impact of different parameters.