Characterizing and Learning Equivalence Classes of Causal DAGs under Interventions

By Karren D. Yang, Abigail Katoff, Caroline Uhler. 2018

135 citations recorded by the source corpus.

We consider the problem of learning causal DAGs in the setting where both observational and interventional data is available. This setting is common in biology, where gene regulatory networks can be intervened on using chemical reagents or gene deletions. Hauser and B\"uhlmann (2012) previously characterized the identifiability of causal DAGs under perfect interventions, which eliminate dependencies between targeted variables and their direct causes. In this paper, we extend these identifiability results to general interventions, which may modify the dependencies between targeted variables and their causes without eliminating them. We define and characterize the interventional Markov equivalence class that can be identified from general (not necessarily perfect) intervention experiments. We also propose the first provably consistent algorithm for learning DAGs in this setting and evaluate our algorithm on simulated and biological datasets.

Subfields: Artificial Intelligence, Computational Theory and Mathematics, Control and Systems Engineering, Causal Inference & Discovery

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computational Theory and Mathematics

In Control and Systems Engineering

In Causal Inference & Discovery

Citations in the corpus

Differentiable Causal Discovery from Interventional Data (2020, methodology): “A specific case that is often considered is (stochastic) perfect interventions (or hard, structural) [10, 48, 26] where p j (xj |xπG j ) = p (k) j (xj) for all j ∈ Ik, thus removing the dependencies with their parents (see Figure 1).”
D’ya Like DAGs? A Survey on Structure Learning and Causal Discovery (2021, unknown): “PC [223] 1993 constraint yes yes yes no CPDAG CCD [196] 1996 constraint yes yes both no PAG FCI [223] 2000 constraint no yes yes no PAG TPDA [31] 2002 constraint yes yes yes no CPDAG CPC [190] 2006 constraint yes relaxed yes no CPDAG KCL [227] 2007 constraint yes yes yes no CPDAG ION [234] 2008 cons”
Permutation-Based Causal Structure Learning with Unknown Intervention Targets (2019, methodology): “One approach for learning the I -MEC is to ﬁrst estimate { ˆ I k } k ∈ [ K ] , and then apply algorithms such as IGSP (Yang et al., 2018) that operates in the setting where the intervention targets are known.”
Joint Causal Inference from Multiple Contexts (2016, methodology): “, 2017) - + - + + + + + + + - b (Yang et al., 2018) - + - + + - - - - + - a/b (Forré and Mooij, 2018) + + + + - - - - - + + a Joint Causal Inference (this work) + + + + + + + + + + - b FCI-JCI (this work) + + ? + + + + + + + - b ASD-JCI (this work) + + + + + + + + + + - b”
Characterization and Learning of Causal Graphs with Latent Variables from Soft Interventions (2019, background): “More recently, [24] showed that the same characterization can be used for both hard and soft interventions.”
Causal Discovery from Soft Interventions with Unknown Targets: Characterization and Learning (2020, background): “The works in [9, 34, 17] characterize the so called I-Markov equivalence class, which uses distributional invariances within and across the available mixture of observational and interventional distributions.”