Differentiable Causal Discovery from Interventional Data

By Philippe Brouillard, Sébastien Lachapelle, Alexandre Lacoste, Simon Lacoste-Julien, Alexandre Drouin. 2020

282 citations recorded by the source corpus.

Discovering causal relationships in data is a challenging task that involves solving a combinatorial problem for which the solution is not always identifiable. A new line of work reformulates the combinatorial problem as a continuous constrained optimization one, enabling the use of different powerful optimization techniques. However, methods based on this idea do not yet make use of interventional data, which can significantly alleviate identifiability issues. In this work, we propose a neural network-based method for this task that can leverage interventional data. We illustrate the flexibility of the continuous-constrained framework by taking advantage of expressive neural architectures such as normalizing flows. We show that our approach compares favorably to the state of the art in a variety of settings, including perfect and imperfect interventions for which the targeted nodes may even be unknown.

Subfields: Artificial Intelligence, 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 Causal Inference & Discovery

Citations in the corpus

Bayesian Structure Learning with Generative Flow Networks (2022, neutral): “Note that this setup is different from previous works which use continuous data instead (Brouillard et al., 2020; Faria et al., 2022).”
Disentanglement via Mechanism Sparsity Regularization: A New Principle for Nonlinear ICA (2021, methodology): “…has been explored previously in the context of causal discovery where the intervention occurs on observed variables instead of latent variables like in our case (Eaton and Murphy, 2007; Mooij et al., 2020; Squires et al., 2020; Jaber et al., 2020; Brouillard et al., 2020; Ke et al., 2019).”
Differentiable Causal Discovery Under Latent Interventions (2022, background): “HD tp fn fp rev F 1 score GIES (Hauser and B¨uhlmann, 2012) 38 10 0 41 7 0.33 CAM (B¨uhlmann et al., 2014) 35 12 1 30 4 0.51 IGSP (Wang et al., 2017) 18 4 6 5 7 0.42 DCDI-G (Brouillard et al., 2020) 36 6 2 25 9 0.31 DCDI-DSF (Brouillard et al., 2020) 2020), with the exception of FCI.”
DiBS: Differentiable Bayesian Structure Learning (2021, background): “Follow-up work has leveraged this insight to model nonlinear relationships [24–27], time-series data [28], in the context of generative modeling [29, 30], for causal inference [31–33], and contributed to its theoretical understanding [34, 35].”