I-FLOP: Fast Learning of Order and Parents from Interventional Data

Liuting Chen, Alex Markham
Proceedings of the 13th International Conference on Probabilistic Graphical Models, PMLR 343:246-273, 2026.

Abstract

We extend the FLOP (fast learning of order and parents) algorithm recently developed by Wienöbst et al. (2026) from observational to interventional data. In particular, we use the interventional BIC score of Hauser and Bühlmann (2012), adapting it to be used with the iterative Cholesky-based score updates that are partly responsible for FLOP’s speed. We show that, in the sample limit, I-FLOP recovers a DAG in the same interventional Markov equivalence class as the data-generating DAG. We compare I-FLOP to existing causal structure learning algorithms on simulated and real interventional data, where it performs favorably in terms of both performance and run time.

Cite this Paper

BibTeX
@InProceedings{pmlr-v343-chen26a, title = {{I-FLOP}: Fast Learning of Order and Parents from Interventional Data}, author = {Chen, Liuting and Markham, Alex}, booktitle = {Proceedings of the 13th International Conference on Probabilistic Graphical Models}, pages = {246--273}, year = {2026}, editor = {Camps-Valls, Gustau and Leonelli, Manuele and Varando, Gherardo}, volume = {343}, series = {Proceedings of Machine Learning Research}, month = {09--11 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v343/main/assets/chen26a/chen26a.pdf}, url = {https://proceedings.mlr.press/v343/chen26a.html}, abstract = {We extend the FLOP (fast learning of order and parents) algorithm recently developed by Wienöbst et al. (2026) from observational to interventional data. In particular, we use the interventional BIC score of Hauser and Bühlmann (2012), adapting it to be used with the iterative Cholesky-based score updates that are partly responsible for FLOP’s speed. We show that, in the sample limit, I-FLOP recovers a DAG in the same interventional Markov equivalence class as the data-generating DAG. We compare I-FLOP to existing causal structure learning algorithms on simulated and real interventional data, where it performs favorably in terms of both performance and run time.} }
Endnote
%0 Conference Paper %T I-FLOP: Fast Learning of Order and Parents from Interventional Data %A Liuting Chen %A Alex Markham %B Proceedings of the 13th International Conference on Probabilistic Graphical Models %C Proceedings of Machine Learning Research %D 2026 %E Gustau Camps-Valls %E Manuele Leonelli %E Gherardo Varando %F pmlr-v343-chen26a %I PMLR %P 246--273 %U https://proceedings.mlr.press/v343/chen26a.html %V 343 %X We extend the FLOP (fast learning of order and parents) algorithm recently developed by Wienöbst et al. (2026) from observational to interventional data. In particular, we use the interventional BIC score of Hauser and Bühlmann (2012), adapting it to be used with the iterative Cholesky-based score updates that are partly responsible for FLOP’s speed. We show that, in the sample limit, I-FLOP recovers a DAG in the same interventional Markov equivalence class as the data-generating DAG. We compare I-FLOP to existing causal structure learning algorithms on simulated and real interventional data, where it performs favorably in terms of both performance and run time.
APA
Chen, L. & Markham, A.. (2026). I-FLOP: Fast Learning of Order and Parents from Interventional Data. Proceedings of the 13th International Conference on Probabilistic Graphical Models, in Proceedings of Machine Learning Research 343:246-273 Available from https://proceedings.mlr.press/v343/chen26a.html.

Related Material