Diffusion models from scratch, from a new theoretical perspective

Re:Diffusion models from scratch, from a new theoretical per... by eutropia on Tuesday March 12, 2024 @12:41PM (#39681602)

In your example images at the very end, the momentum term seems to have a deleterious effect on the digital painting of the house (the door is missing in the gamma = 2.0 image). I would like to know more details of that example to build an intuition for the effect of your gradient-informed ddim sampler.

As someone who's spent a little time experimenting with sampling procedures on stable diffusion, I was also hoping for a comparison to DDIM in terms of convergence time/steps. Is there a relationship between momentum, convergence and error? (i.e, your momentum sampler at 16 steps is ~equivalent to ddim at 20 steps ± %err_term)

thanks for the excellent post

[ Parent | on HN ]

Re:Diffusion models from scratch, from a new theoretical per... by ycy on Tuesday March 12, 2024 @01:04PM (#39681971)

It is hard to quantify the performance of samplers on individual images, but we have quantitative experiments on pretrained models in our paper (https://arxiv.org/pdf/2306.04848.pdf). Table 2 has a comparison to other samplers and Figure 9 plots the effects of the momentum term on convergence.

[ Parent | on HN ]