Dagma-DCE: Interpretable, Non-Parametric Differentiable Causal Discovery
In: IEEE Open Journal of Signal Processing, vol. 5, pp. 393-401, 2024; (2024) S. 393-401
Online
report
We introduce Dagma-DCE, an interpretable and model-agnostic scheme for differentiable causal discovery. Current non- or over-parametric methods in differentiable causal discovery use opaque proxies of ``independence'' to justify the inclusion or exclusion of a causal relationship. We show theoretically and empirically that these proxies may be arbitrarily different than the actual causal strength. Juxtaposed to existing differentiable causal discovery algorithms, \textsc{Dagma-DCE} uses an interpretable measure of causal strength to define weighted adjacency matrices. In a number of simulated datasets, we show our method achieves state-of-the-art level performance. We additionally show that \textsc{Dagma-DCE} allows for principled thresholding and sparsity penalties by domain-experts. The code for our method is available open-source at https://github.com/DanWaxman/DAGMA-DCE, and can easily be adapted to arbitrary differentiable models.
Comment: 9 pages, 2 figures. Accepted to the IEEE Open Journal of Signal Processing
Titel: |
Dagma-DCE: Interpretable, Non-Parametric Differentiable Causal Discovery
|
---|---|
Autor/in / Beteiligte Person: | Waxman, Daniel ; Butler, Kurt ; Djuric, Petar M. |
Link: | |
Quelle: | IEEE Open Journal of Signal Processing, vol. 5, pp. 393-401, 2024; (2024) S. 393-401 |
Veröffentlichung: | 2024 |
Medientyp: | report |
DOI: | 10.1109/OJSP.2024.3351593 |
Schlagwort: |
|
Sonstiges: |
|