Truncated Diffusion Probabilistic Models and Diffusion-based Adversarial Auto-Encoders
Huangjie Zheng1,2、Pengcheng He2、Weizhu Chen2、Mingyuan Zhou1 The University of Texas at Austin1, Microsoft Azure AI2
Abstract
Employing a forward diffusion chain to gradually map the data to a noise distribution, diffusion-based generative models learn how to generate the data by inferring a reverse diffusion chain. However, this approach is slow and costly because it needs many forward and reverse steps. We propose a faster and cheaper approach that adds noise not until the data become pure random noise, but until they reach a hidden noisy-data distribution that we can confidently learn. Then, we use fewer reverse steps to generate data by starting from this hidden distribution that is made similar to the noisy data. We reveal that the proposed model can be cast as an adversarial auto-encoder empowered by both the diffusion process and a learnable implicit prior. Experimental results show even with a significantly smaller number of reverse diffusion steps, the proposed truncated diffusion probabilistic models can provide consistent improvements over the non-truncated ones in terms of performance in both unconditional and text-guided image generations.
中文速览
扩散模型（diffusion model）生成图像效果很好，但需要走几百甚至几千步正向加噪、再一步步反向去噪，速度慢、算力贵。本文提出"截断扩散概率模型"（Truncated Diffusion Probabilistic Model, TDPM），核心思路是：不把数据一路加噪到纯高斯白噪声，而是只走前面一小段，让数据停在一个"半噪声"的隐分布上，再用一个可学习的隐式生成器（implicit generator）来模拟这个分布作为反向过程的起点，从而大幅减少去噪步数。理论上，作者证明这套框架等价于一个由扩散过程赋能的对抗自编码器（Adversarial Auto-Encoder, AAE），隐式先验由对抗训练学得，而非人工指定。实验表明，TDPM 在无条件图像生成上以更少的步数持续超越原版 DDPM，其文本引导版本 TLDM 在同等质量下生成速度比潜在扩散模型（Latent Diffusion Model, LDM）快 50 倍以上，为在资源受限场景下部署高质量扩散模型提供了切实可行的路径。
原文 arXiv:2202.09671;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2202.09671v4