论文标题
分析-DPM：扩散概率模型中最佳反向差异的分析估计值
Analytic-DPM: an Analytic Estimate of the Optimal Reverse Variance in Diffusion Probabilistic Models
论文作者
论文摘要
扩散概率模型（DPM）代表一类强大的生成模型。尽管他们成功了，但DPM的推断还是很昂贵的，因为它通常需要迭代数千个时间段。推断中的一个关键问题是估计反向过程每个时间步的差异。在这项工作中，我们提出了一个令人惊讶的结果，即DPM的最佳反向差异和相应的最佳KL差异具有分析形式W.R.T.它的得分功能。在此基础上，我们建议使用Monte Carlo方法和基于得分的模型来估计方差和KL差异的分析形式的Analytic-DPM。此外，为了纠正基于分数模型引起的潜在偏差，我们得出了最佳方差的下层和上限，并夹住估计值以获得更好的结果。从经验上讲，我们的分析DPM改善了各种DPM的对数可能性，产生高质量的样本，同时享有20倍至80倍的速度。
Diffusion probabilistic models (DPMs) represent a class of powerful generative models. Despite their success, the inference of DPMs is expensive since it generally needs to iterate over thousands of timesteps. A key problem in the inference is to estimate the variance in each timestep of the reverse process. In this work, we present a surprising result that both the optimal reverse variance and the corresponding optimal KL divergence of a DPM have analytic forms w.r.t. its score function. Building upon it, we propose Analytic-DPM, a training-free inference framework that estimates the analytic forms of the variance and KL divergence using the Monte Carlo method and a pretrained score-based model. Further, to correct the potential bias caused by the score-based model, we derive both lower and upper bounds of the optimal variance and clip the estimate for a better result. Empirically, our analytic-DPM improves the log-likelihood of various DPMs, produces high-quality samples, and meanwhile enjoys a 20x to 80x speed up.
