Causal Graph Learning via Distributional Invariance of Cause-Effect Relationship
Nang Hung Nguyen [email protected] · 2026
Read Causal Graph Learning via Distributional Invariance of Cause-Effect Relationship by Nang Hung Nguyen [email protected] (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-2602-03353.txt.
Description
Causal graphs are recovered from an invariance property by Nang Hung Nguyen, leading 4 other authors, in a system exploiting distributional stability. The spread of an effect, conditioned on its causes, stays invariant when the prior spread of those causes changes. That observation yields a direct test for potential causal relationships. It checks how much the effect-cause conditional spread varies across data subsets. Those subsets reflect other cause priors while keeping the conditional relationship itself.
Combined with the empirical sparsity of most causal graphs, the test becomes an efficient learning method. Its complexity is quadratic in the number of variables, cutting processing time dramatically against leading methods. Tests on a varied benchmark of large-scale datasets showed superior or equivalent results. The authors emphasize scalability to large variable sets as the main advance.
Catalog record
| CREATOR | Nang Hung Nguyen [email protected] |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 111 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-03353