PreprintCC BY-NC-SA 4.0arXiv
I-FLOP: Fast Learning of Order and Parents from Interventional Data
Abstract
We extend the FLOP (fast learning of order and parents) algorithm recently proposed 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 real and simulated interventional data, where it performs favorably in terms of both performance and run time.
Related
- Gromov-Witten theory of the local model of the standard flop
- Degenerations of exotic Calabi-Yau metrics through Atiyah's flop
- Fast Rates and Strong Convergence of Tikhonov-Regularized Mixed-Order Primal-Dual Dynamics for Linearly Constrained Optimization Without Eventual Ball Conditions
- Slow--Fast Response Correction for a Compact Fourth-Order IMEX Two-Derivative Method
- Cartan-Fejer Gram Tomography and Flop Covariance for BPS Resummed Gromov-Witten Potentials
BibTeX
@misc{chen2026iflop,
title = {{I-FLOP: Fast Learning of Order and Parents from Interventional Data}},
author = {Liuting Chen and Alex Markham},
howpublished = {arXiv},
year = {2026},
month = aug,
doi = {10.48550/arxiv.2608.28245},
eprint = {2608.28245},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2608.28245},
note = {CC BY-NC-SA 4.0},
}
RIS
TY - UNPB
TI - I-FLOP: Fast Learning of Order and Parents from Interventional Data
AU - Liuting Chen
AU - Alex Markham
PY - 2026
DA - 2026/08/28
PB - arXiv
DO - 10.48550/arxiv.2608.28245
UR - https://arxiv.org/abs/2608.28245
AB - We extend the FLOP (fast learning of order and parents) algorithm recently proposed 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 real and simulated interventional data, where it performs favorably in terms of both performance and run time.
ER -