@YiMaTweets Exactly, in our recent papers, we connected the channel coding theory with DL, in which we visualized the learning process, and figure out the semantics meaning of information theory quantities. It could be a first step towards theoretically sound DL.
https://t.co/tuDVuORU7j
Generating ~200 million parameters in just minutes! 🥳
Excited to share our work with @MTDovent , @heisejiasuo96 , and @YangYou1991: 'Recurrent Diffusion for Large-Scale Parameter Generation' (RPG for short).
Example: Obtain customized models using prompts (see below). (🧵1/8)