Enes Causal Discovery
Alexis Kafantaris · 2026
Read Enes Causal Discovery by Alexis Kafantaris (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-2603-24436.txt.
Description
Kafantaris proposes Enes, a causal discovery architecture built as a mixture of experts. The design lets model entities such as causal relationships be parameterised further, though fitting a neural net to this dataset proves hard. A simple, fast Pearson coefficient linear model already scores well, setting a tough baseline to beat. The work also stresses the limits of causal discovery from observational data alone. Unlike the Sachs study, no interventions are used, only prior knowledge, and the data itself is a key constraint.
Catalog record
| CREATOR | Alexis Kafantaris |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 15974 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.24436