Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2311.17101 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910536099168256 |
|---|---|
| author | Dao, Quan Ta, Binh Pham, Tung Tran, Anh |
| author_facet | Dao, Quan Ta, Binh Pham, Tung Tran, Anh |
| contents | Developing image-generative models, which are robust to outliers in the training process, has recently drawn attention from the research community. Due to the ease of integrating unbalanced optimal transport (UOT) into adversarial framework, existing works focus mainly on developing robust frameworks for generative adversarial model (GAN). Meanwhile, diffusion models have recently dominated GAN in various tasks and datasets. However, according to our knowledge, none of them are robust to corrupted datasets. Motivated by DDGAN, our work introduces the first robust-to-outlier diffusion. We suggest replacing the UOT-based generative model for GAN in DDGAN to learn the backward diffusion process. Additionally, we demonstrate that the Lipschitz property of divergence in our framework contributes to more stable training convergence. Remarkably, our method not only exhibits robustness to corrupted datasets but also achieves superior performance on clean datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2311_17101 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | A High-Quality Robust Diffusion Framework for Corrupted Dataset Dao, Quan Ta, Binh Pham, Tung Tran, Anh Computer Vision and Pattern Recognition Developing image-generative models, which are robust to outliers in the training process, has recently drawn attention from the research community. Due to the ease of integrating unbalanced optimal transport (UOT) into adversarial framework, existing works focus mainly on developing robust frameworks for generative adversarial model (GAN). Meanwhile, diffusion models have recently dominated GAN in various tasks and datasets. However, according to our knowledge, none of them are robust to corrupted datasets. Motivated by DDGAN, our work introduces the first robust-to-outlier diffusion. We suggest replacing the UOT-based generative model for GAN in DDGAN to learn the backward diffusion process. Additionally, we demonstrate that the Lipschitz property of divergence in our framework contributes to more stable training convergence. Remarkably, our method not only exhibits robustness to corrupted datasets but also achieves superior performance on clean datasets. |
| title | A High-Quality Robust Diffusion Framework for Corrupted Dataset |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2311.17101 |