Revisit Tabular Data Generator through Differential Privacy Perspective
- 2024-08-12 (Mon.), 10:30 AM
- Auditorium, B1F, Institute of Statistical Science；The tea reception will be held at 10:10.
- Lecture in English. Online live streaming through Cisco Webex will be available.
- Dr. Chi-Hua Wang
- Department of Statistics and Data Science, University of California, Los Angeles (UCLA)
Abstract
In this talk, we will walk through 3 representative tabular data generator from the eye of Differential Privacy. These tabular data generator are CTGAN (GAN-based), TabDDPM (Diffusion-based) and GReaT (Language-based). We will first start from a digital advertisement use case to motivate the task of tabular data generation. Then we will discuss specific challenges regards to each tabular data generator. Then we will motivate the differential privacy of synthetic data from the task of data sharing and review the differential privacy notion in sharing unstructured data.
Please click here for participating the talk online.
Download
Update：2024-08-05 14:42
Back
