A Function-Space Approach to the Statistical Mechanics of Learning Dynamics
This story is from 2026-09-11. It is preserved in the archive; the latest stories are on the live feed.
arXiv:2609.09589v1 Announce Type: new Abstract: Deep neural networks exhibit regular macroscopic behavior despite highly nonlinear dynamics in vast parameter spaces. We develop a statistical-mechanical description of learning directly in function space, treating parameter configurations as microsco…
Read the full story at arXiv cs.AI ↗
Timeline · 1 report
- 2026-09-11 04:00 · arXiv cs.AI
A Function-Space Approach to the Statistical Mechanics of Learning Dynamics