Leveraging Programmatically Generated Synthetic Data for Differentially Private Diffusion Training
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913610630955008 |
|---|---|
| author | Choi, Yujin Park, Jinseong Byun, Junyoung Lee, Jaewook |
| author_facet | Choi, Yujin Park, Jinseong Byun, Junyoung Lee, Jaewook |
| contents | Programmatically generated synthetic data has been used in differential private training for classification to enhance performance without privacy leakage. However, as the synthetic data is generated from a random process, the distribution of real data and the synthetic data are distinguishable and difficult to transfer. Therefore, the model trained with the synthetic data generates unrealistic random images, raising challenges to adapt the synthetic data for generative models. In this work, we propose DP-SynGen, which leverages programmatically generated synthetic data in diffusion models to address this challenge. By exploiting the three stages of diffusion models(coarse, context, and cleaning) we identify stages where synthetic data can be effectively utilized. We theoretically and empirically verified that cleaning and coarse stages can be trained without private data, replacing them with synthetic data to reduce the privacy budget. The experimental results show that DP-SynGen improves the quality of generative data by mitigating the negative impact of privacy-induced noise on the generation process. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_09842 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Leveraging Programmatically Generated Synthetic Data for Differentially Private Diffusion Training Choi, Yujin Park, Jinseong Byun, Junyoung Lee, Jaewook Machine Learning Computer Vision and Pattern Recognition Programmatically generated synthetic data has been used in differential private training for classification to enhance performance without privacy leakage. However, as the synthetic data is generated from a random process, the distribution of real data and the synthetic data are distinguishable and difficult to transfer. Therefore, the model trained with the synthetic data generates unrealistic random images, raising challenges to adapt the synthetic data for generative models. In this work, we propose DP-SynGen, which leverages programmatically generated synthetic data in diffusion models to address this challenge. By exploiting the three stages of diffusion models(coarse, context, and cleaning) we identify stages where synthetic data can be effectively utilized. We theoretically and empirically verified that cleaning and coarse stages can be trained without private data, replacing them with synthetic data to reduce the privacy budget. The experimental results show that DP-SynGen improves the quality of generative data by mitigating the negative impact of privacy-induced noise on the generation process. |
| title | Leveraging Programmatically Generated Synthetic Data for Differentially Private Diffusion Training |
| topic | Machine Learning Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2412.09842 |