Exploring the Dimensions of a Variational Neuron
Yves Ruffenach · 2026
Read Exploring the Dimensions of a Variational Neuron by Yves Ruffenach (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-2603-13849.txt.
Description
Ruffenach introduces EVE, a neuron that carries its own probability distribution instead of a single number. Each unit has an explicit prior, a learned posterior, and its own regularization term. The study varies the neuron's inner dimension, from one up to larger latent spaces, and tracks how its working regime changes. Two added axes, local capacity control and a time-based extension, are also tested. Diagnostics show some inner variables can be measured and link to downstream behavior in forecasting and tabular tasks.
Catalog record
| CREATOR | Yves Ruffenach |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 27995 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.13849