diffGHOST: Diffusion based Generative Hedged Oblivious Synthetic Trajectories
For researchers and practitioners in mobility data analysis, this work provides a method to generate privacy-preserving synthetic trajectories, but the novelty is incremental as it adapts existing diffusion models with a segmentation technique.
The paper introduces diffGHOST, a conditional diffusion model with latent space segmentation that generates synthetic trajectories with formal privacy guarantees while preserving utility, addressing the failure of prior generative models to provide privacy guarantees.
Trajectories are nowadays valuable information for a wide range of applications. However they are also inherently sensitive, as they contain highly personal information about individuals. Facing this challenge, synthesizing mobility trajectories has emerged as a promising solution to leverage mobility information while preserving privacy. State-of-the-art models, often rely on the false assumptions of generative models implicit privacy and fails to provide privacy guarantees while preserving trajectories utility. Here, we introduce diffGHOST, a conditional diffusion model based on latent space segmentation, designed to answer this challenge. Thus, this paper propose a methodology that identify and mitigate memorization of critical samples using condition segments of a learn latent space.