EDC: Equation Discovery for Classification
Guus Toussaint, Arno Knobbe · 2025
Read EDC: Equation Discovery for Classification by Guus Toussaint, Arno Knobbe (2025) 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-2510-24310.txt.
Description
Guus Toussaint and Arno Knobbe extend a search method that finds short, clear formulas from data. In the past, this search served tasks where the output is a number. Their new method, EDC, serves yes or no tasks. It looks for a short analytic formula that states where the border between the two classes lies and what shape it has. That keeps the learned model easy to read and check.
On made up data and on real data, EDC found both the true form of the target rule and the values of its constants. It beat past formula search methods for class tasks. Its scores were on par with the best class tools overall. The authors suggest a grammar to guide the search and keep the found borders small and clear.
Catalog record
| CREATOR | Guus Toussaint, Arno Knobbe |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2510-24310