ComboStoc: Combinatorial Stochasticity for Diffusion Generative Models
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911631088287744 |
|---|---|
| author | Xu, Rui Wang, Jiepeng Pan, Hao Liu, Yang Tong, Xin Xin, Shiqing Tu, Changhe Komura, Taku Wang, Wenping |
| author_facet | Xu, Rui Wang, Jiepeng Pan, Hao Liu, Yang Tong, Xin Xin, Shiqing Tu, Changhe Komura, Taku Wang, Wenping |
| contents | In this paper, we study an under-explored but important factor of diffusion generative models, i.e., the combinatorial complexity. Data samples are generally high-dimensional, and for various structured generation tasks, additional attributes are combined to associate with data samples. We show that the space spanned by the combination of dimensions and attributes can be insufficiently covered by existing training schemes of diffusion generative models, potentially limiting test time performance. We present a simple fix to this problem by constructing stochastic processes that fully exploit the combinatorial structures, hence the name ComboStoc. Using this simple strategy, we show that network training is significantly accelerated across diverse data modalities, including images and 3D structured shapes. Moreover, ComboStoc enables a new way of test time generation which uses asynchronous time steps for different dimensions and attributes, thus allowing for varying degrees of control over them. Our code is available at: https://github.com/Xrvitd/ComboStoc |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_13729 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | ComboStoc: Combinatorial Stochasticity for Diffusion Generative Models Xu, Rui Wang, Jiepeng Pan, Hao Liu, Yang Tong, Xin Xin, Shiqing Tu, Changhe Komura, Taku Wang, Wenping Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition Graphics In this paper, we study an under-explored but important factor of diffusion generative models, i.e., the combinatorial complexity. Data samples are generally high-dimensional, and for various structured generation tasks, additional attributes are combined to associate with data samples. We show that the space spanned by the combination of dimensions and attributes can be insufficiently covered by existing training schemes of diffusion generative models, potentially limiting test time performance. We present a simple fix to this problem by constructing stochastic processes that fully exploit the combinatorial structures, hence the name ComboStoc. Using this simple strategy, we show that network training is significantly accelerated across diverse data modalities, including images and 3D structured shapes. Moreover, ComboStoc enables a new way of test time generation which uses asynchronous time steps for different dimensions and attributes, thus allowing for varying degrees of control over them. Our code is available at: https://github.com/Xrvitd/ComboStoc |
| title | ComboStoc: Combinatorial Stochasticity for Diffusion Generative Models |
| topic | Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition Graphics |
| url | https://arxiv.org/abs/2405.13729 |