Enhancing classification accuracy through chaos
Panos Stinis · 2026
Read Enhancing classification accuracy through chaos by Panos Stinis (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-2603-15299.txt.
Description
Stinis proposes a classifier that runs its input through a chaotic system before deciding. Data vectors are lifted into a higher-dimensional space and used as starting states. A chaotic dynamical system then evolves them for a fixed time span. The evolved state feeds a trainable softmax classifier. On perturbed orthogonal vectors with up to twenty classes, the method trained faster and classified better than plain softmax baselines. He also explains the gain and gives a way to pick the best evolution time.
Catalog record
| CREATOR | Panos Stinis |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38494 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.15299