Learning the Integral of a Diffusion Model
21–26 of 26 posts
Re: Learning the Integral of a Diffusion Model
#22- your links to the slides for deeplearning you did here https://sander.ai/2014/05/29/slides-meetup.html are broken
Re: Learning the Integral of a Diffusion Model
#23This is way outside of my expertise, can anyone given a TL;DR or ai;dr?
Diffusion and flow matching models generate samples by iterative denoising. Iterative denoising means passing input to the neural network, running a forward pass, and taking the output back as input and rerunning the neural network. Often you do this 100 times, which is slow and expensive. Flow maps / consistency models / shortcut models instead try to learn to compress this iterative work into 1 forward pass. This m…
Re: Learning the Integral of a Diffusion Model
#24Earlier quoted context omitted.
Why not put it into an AI yourself? :) I'd rather we avoided a precedent of asking for it and N people replying with their own favorite AI version. The comments section would end up a ghost town. Extreme TL;DR: Diffusion models are like getting f(x) by calculating and summing f'(0), f'(1)...f'(x). Flow models are like just calculating f(x).
We've all seen that AI can give you plausible but incorrect answers. Having an expert read it or use AI on it and interpret and validate it before posting would be most welcome IMO.
Re: Learning the Integral of a Diffusion Model
#25Does anyone have good resources into a more practical approach toward building diffusion models? I found the book by Rashka for Building an LLM from Scratch really helpful in understanding a lot of concepts behind LLMs, and I am looking for a similar resource for diffusion models
MIT's OCW is usually pretty reliable: https://www.practical-diffusion.org/lectures/ There is more math-heavy https://diffusion.csail.mit.edu/2026/index.html
Re: Learning the Integral of a Diffusion Model
#26- just a headsup - your links to the slides for deeplearning you did here https://sander.ai/2014/05/29/slides-meetup.html are broken