Incentives in Private Collaborative Machine Learning

arXiv:2404.01676v1 Announce Type: new
Abstract: Collaborative machine learning involves training models on data from multiple parties but must incentivize their participation. Existing data valuation methods fairly value and reward each party based on shared data or model parameters but neglect the privacy risks involved. To address this, we introduce differential privacy (DP) as an incentive. Each party can select its required DP guarantee and perturb its sufficient statistic (SS) accordingly. The mediator values the perturbed SS by the Bayesian surprise …

abstract arxiv collaborative cs.lg data differential differential privacy incentives machine machine learning multiple parameters parties privacy risks training training models type valuation value