ArXiv TLDR

diffGHOST: Diffusion based Generative Hedged Oblivious Synthetic Trajectories

🐦 Tweet
2605.10647

Florent Guépin, Cheick Tidiani Cisse, Denis Renaud, François Bidet, Arnaud Legendre

cs.AIcs.CR

TLDR

diffGHOST is a diffusion model that generates privacy-preserving synthetic mobility trajectories by mitigating memorization in a segmented latent space.

Key contributions

  • Introduces diffGHOST, a novel conditional diffusion model for synthetic trajectory generation.
  • Employs latent space segmentation to enhance privacy protection.
  • Identifies and mitigates memorization of critical sensitive samples.
  • Aims to provide privacy guarantees while preserving trajectory utility.

Why it matters

Mobility trajectories are sensitive, and existing models often lack privacy guarantees. diffGHOST introduces a conditional diffusion model with latent space segmentation to prevent memorization, ensuring privacy-preserving synthetic data.

Original Abstract

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.

📬 Weekly AI Paper Digest

Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.