Privacy Evaluation of Generative Models for Trajectory Generation

TL;DR

This paper examines privacy concerns in generative models for trajectory data, demonstrating that these models do not inherently guarantee privacy and highlighting the need for empirical privacy evaluation methods.

Contribution

It identifies a gap in privacy evaluation for generative trajectory models and demonstrates the effectiveness of membership inference attacks in exposing privacy risks.

Findings

Membership inference attacks can successfully compromise privacy in generative trajectory models.

Generative models do not inherently guarantee privacy preservation.

There is a significant gap in empirical privacy evaluation methods for these models.

Abstract

Trajectory data is fundamental to modern urban intelligence, yet its sensitivity raises significant privacy concerns. Generative models such as Generative Adversarial Networks, Variational Autoencoders, and Diffusion Models have been developed to generate realistic synthetic trajectory data by capturing underlying spatiotemporal distributions and mobility patterns. Although these models are often assumed to preserve privacy due to their generative nature, this assumption does not necessarily hold. In this work, we investigate the intersection of generative trajectory modeling and privacy evaluation. By identifying applicable empirical methods for assessing privacy preservation in trajectory generation tasks, we demonstrate a significant gap in the evaluation of privacy for generative trajectory models. Motivated by this gap, we implement Membership Inference Attacks against representative models, demonstrating the feasibility of using such empirical privacy evaluation methods and showing that their generative nature does not eliminate privacy risks.