Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Permutation-Based Causal Structure Learning with Unknown Intervention Targets

About

We consider the problem of estimating causal DAG models from a mix of observational and interventional data, when the intervention targets are partially or completely unknown. This problem is highly relevant for example in genomics, since gene knockout technologies are known to have off-target effects. We characterize the interventional Markov equivalence class of DAGs that can be identified from interventional data with unknown intervention targets. In addition, we propose a provably consistent algorithm for learning the interventional Markov equivalence class from such data. The proposed algorithm greedily searches over the space of permutations to minimize a novel score function. The algorithm is nonparametric, which is particularly important for applications to genomics, where the relationships between variables are often non-linear and the distribution non-Gaussian. We demonstrate the performance of our algorithm on synthetic and biological datasets. Links to an implementation of our algorithm and to a reproducible code base for our experiments can be found at https://uhlerlab.github.io/causaldag/utigsp.

Chandler Squires, Yuhao Wang, Caroline Uhler• 2019

Related benchmarks

TaskDatasetResultRank
Graph RecoverySynthetic Nonlinear SEM Gaussian Noise (test)
AUPRC100
15
Graph RecoverySynthetic Nonlinear SEM (Gumbel Noise) (test)
AUPRC0.995
15
Graph RecoverySynthetic Nonlinear SEM Exponential Noise (test)
AUPRC92.4
15
Causal DiscoveryNonlinear SCM 1.0 (synthetic)
F1 (NL1)49.5
9
Causal DiscoverySynthetic Linear SCMs Fork, Chain, V-str, Diamond, Collider (test)
Fork Score0.448
9
Causal DiscoveryCausalDynamics Rössler
AUROC45
9
Causal DiscoveryCausalDynamics Lorenz
AUROC54
9
Causal DiscoveryTigramite S6 Latent Variables
F1 Score72
8
Causal DiscoveryCausalDynamics CoupledLorenz
AUROC49.7
8
Causal DiscoveryCausalDynamics CoupledRösslerL.
AUROC58
8
Showing 10 of 22 rows

Other info

Follow for update