Diffusion models from scratch, from a new theoretical perspective

Re:Diffusion models from scratch, from a new theoretical per... by adamnemecek on Monday March 11, 2024 @06:41PM (#39674036)

All machine learning models are convolutions, mark my words.

[ Parent | on HN ]

Re:Diffusion models from scratch, from a new theoretical per... by sva_ on Monday March 11, 2024 @07:33PM (#39674413)

I think you posted this a few times, maybe you want to elaborate on this point?

I have trouble seeing reinforcement learning as convolution, for example.

[ Parent | on HN ]

Re:Diffusion models from scratch, from a new theoretical per... by adamnemecek on Monday March 11, 2024 @08:37PM (#39674796)

It's a Fredholm equation which in turn is a convolutional equation.

[ Parent | on HN ]

Re:Diffusion models from scratch, from a new theoretical per... by carlthome on Monday March 11, 2024 @09:21PM (#39675136)

Next you're gonna say that polynomial multiplication is also just convolution of the arguments.

[ Parent | on HN ]

Re:Diffusion models from scratch, from a new theoretical per... by adamnemecek on Monday March 11, 2024 @09:31PM (#39675194)

It's not the standard convolution.

[ Parent | on HN ]