Modern deep learning is very complex. We are hopeful that many lines of research will improve our scientific understanding of DL through the lens of learning mechanics. Hope to find more solveable models & limits, empirical laws, scaling arguments, universal phenomena, etc
1/ Deep learning is going to have a scientific theory. We can see the pieces starting to come together, and it's looking a lot like physics!
We're releasing a paper pulling together these emerging threads and giving them a name: learning mechanics.
🔨
arxiv.org/pdf/2604.21691 🔧