Thanks for watching everyone! If you think I deserve it, please consider giving this video a like. Architecture diagrams in the GitHub link in the description of the video
Diffusion Models are very interesting and their applications are vast. I would love a math-centric explanation of various concepts such as DDPM, DDIM Consistency Models and especially more recent hot topics such as Normalizing Flows, Diffusion Transformer DiT architecture and practical examples such as Stable Diffusion 3 with its MMDiT modules or the recent FLUX architecture.
The only thing I really want to do with AI is replicate my typing style and online behavior so I can become digitally immortal. I don't really trust anyone else to do it for me either, so I can't just pay a company to do it for me
loved your videos. i am curious about first trivia question you asked. can you provide answer to that please. why input and output shape in transformer are same. i think, its (D).
Thanks for watching everyone! If you think I deserve it, please consider giving this video a like. Architecture diagrams in the GitHub link in the description of the video
Diffusion Models are very interesting and their applications are vast. I would love a math-centric explanation of various concepts such as DDPM, DDIM Consistency Models and especially more recent hot topics such as Normalizing Flows, Diffusion Transformer DiT architecture and practical examples such as Stable Diffusion 3 with its MMDiT modules or the recent FLUX architecture.
definitely love the details of model building and the theory of why a particular model works and on what problems its designed for
The only thing I really want to do with AI is replicate my typing style and online behavior so I can become digitally immortal. I don't really trust anyone else to do it for me either, so I can't just pay a company to do it for me
loved your videos. i am curious about first trivia question you asked. can you provide answer to that please. why input and output shape in transformer are same. i think, its (D).
First comment !! :D
great video
Self-improving systems.
Reinforcement Learning