Topological Sensitivity in Connectome-Constrained Neural Networks
Nalin Dhiman · 2026
Read Topological Sensitivity in Connectome-Constrained Neural Networks by Nalin Dhiman (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-04033.txt.
Description
Nalin Dhiman revisited claims that biological graph topology improves learning in connectome-constrained neural networks. Under weak controls, a Drosophila connectome model looked better than sparse random controls on early loss, activity, and runtime. Training both graphs from a shared random start erased the loss edge, and a degree-preserving rewired null erased the activity edge. The study concludes that earlier topology advantages largely reflect initialization and null-model confounds, not genuine biological benefit.
Catalog record
| CREATOR | Nalin Dhiman |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 34767 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.04033