Symmetry without a manifold: intrinsic dimension on orbits
This story is from 2026-09-17. It is preserved in the archive; the latest stories are on the live feed.
arXiv:2609.17926v1 Announce Type: cross Abstract: The standard geometric derivation of neural scaling exponents takes the intrinsic dimension of a data manifold as its input. On modular addition in $\mathbb{Z}_p$ that derivation has no input. The exact algebraic solution is an orbit of $\mathbb{Z}_…
Read the full story at arXiv stat.ML ↗
Timeline · 1 report
- 2026-09-17 04:00 · arXiv stat.ML
Symmetry without a manifold: intrinsic dimension on orbits