2022-06-28-bao22d.md

title	booktitle	abstract	layout	series	publisher	issn	id	month	tex_title	firstpage	lastpage	page	order	cycles	bibtex_author	author	date	address	container-title	volume	genre	issued	pdf	extras
Estimating the Optimal Covariance with Imperfect Mean in Diffusion Probabilistic Models	Proceedings of the 39th International Conference on Machine Learning	Diffusion probabilistic models (DPMs) are a class of powerful deep generative models (DGMs). Despite their success, the iterative generation process over the full timesteps is much less efficient than other DGMs such as GANs. Thus, the generation performance on a subset of timesteps is crucial, which is greatly influenced by the covariance design in DPMs. In this work, we consider diagonal and full covariances to improve the expressive power of DPMs. We derive the optimal result for such covariances, and then correct it when the mean of DPMs is imperfect. Both the optimal and the corrected ones can be decomposed into terms of conditional expectations over functions of noise. Building upon it, we propose to estimate the optimal covariance and its correction given imperfect mean by learning these conditional expectations. Our method can be applied to DPMs with both discrete and continuous timesteps. We consider the diagonal covariance in our implementation for computational efficiency. For an efficient practical implementation, we adopt a parameter sharing scheme and a two-stage training process. Empirically, our method outperforms a wide variety of covariance design on likelihood results, and improves the sample quality especially on a small number of timesteps.	inproceedings	Proceedings of Machine Learning Research	PMLR	2640-3498	bao22d	0	Estimating the Optimal Covariance with Imperfect Mean in Diffusion Probabilistic Models	1555	1584	1555-1584	1555	false	Bao, Fan and Li, Chongxuan and Sun, Jiacheng and Zhu, Jun and Zhang, Bo	given family Fan Bao given family Chongxuan Li given family Jiacheng Sun given family Jun Zhu given family Bo Zhang	2022-06-28		Proceedings of the 39th International Conference on Machine Learning	162	inproceedings	date-parts 2022 6 28	https://proceedings.mlr.press/v162/bao22d/bao22d.pdf