Parameter Privacy-Preserving Data Sharing: A Particle-Belief MDP Formulation

TL;DR

This paper introduces a particle-belief MDP framework for privacy-preserving data sharing in continuous systems, balancing privacy, utility, and system performance through an efficient, tractable optimization approach.

Contribution

It develops a novel particle-belief MDP formulation with Gaussian mixture approximations for effective privacy-utility trade-offs in continuous data sharing.

Findings

The proposed method effectively impedes inference attacks on sensitive parameters.

It maintains data usability and system performance in a mixed-autonomy platoon.

The particle-belief MDP converges asymptotically with increasing particles.

Abstract

This paper investigates parameter-privacy-preserving data sharing in continuous-state dynamical systems, where a data owner designs a data-sharing policy to support downstream estimation and control while preventing adversarial inference of a sensitive parameter. This data-sharing problem is formulated as an optimization problem that trades off privacy leakage and the impact of data sharing on the data owner's utility, subject to a data-usability constraint. We show that this problem admits an equivalent belief Markov decision process (MDP) formulation, which provides a simplified representation of the optimal policy. To efficiently characterize information-theoretic privacy leakage in continuous state and action spaces, we propose a particle-belief MDP formulation that tracks the parameter posterior via sequential Monte Carlo, yielding a tractable belief-state approximation that converges asymptotically as the number of particles increases. We further derive a tractable closed-form upper bound on particle-based MI via Gaussian mixture approximations, which enables efficient optimization of the particle-belief MDP. Experiments on a mixed-autonomy platoon show that the learned continuous policy substantially impedes inference attacks on human-driving behavior parameters while maintaining data usability and system performance.