Modeling Discrete Interventional Data using Directed Cyclic Graphical Models

By Mark W. Schmidt, Kevin P. Murphy. 2009

43 citations recorded by the source corpus.

We outline a representation for discrete multivariate distributions in terms of interventional potential functions that are globally normalized. This representation can be used to model the effects of interventions, and the independence properties encoded in this model can be represented as a directed graph that allows cycles. In addition to discussing inference and sampling with this representation, we give an exponential family parametrization that allows parameter estimation to be stated as a convex optimization problem; we also give a convex relaxation of the task of simultaneous parameter and structure learning using group l1-regularization. The model is evaluated on simulated data and intracellular flow cytometry data.

Subfields: Artificial Intelligence, Molecular Biology, Statistics and Probability

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

1. Bernhard Schölkopf
2. Dominik Janzing
3. Joris M. Mooij
4. Jan Peters
5. Kun Zhang
6. David Lopez-Paz
7. Peter Buhlmann
8. Yoshua Bengio

In Molecular Biology

1. Joris M. Mooij
2. Bernhard Schölkopf
3. Dominik Janzing
4. Matti Järvisalo
5. Frederick Eberhardt
6. Antti Hyttinen
7. Patrik O. Hoyer
8. Marloes H. Maathuis

In Statistics and Probability

1. Bernhard Schölkopf
2. Dominik Janzing
3. Joris M. Mooij
4. Jan Peters
5. Patrik O. Hoyer
6. Frederick Eberhardt
7. Antti Hyttinen
8. David Lopez-Paz

Citations in the corpus

• Discovering Cyclic Causal Models with Latent Variables: A General SAT-Based Procedure (2013, background): “In recent years, many approaches to this causal discovery problem have been suggested (Spirtes et al., 1999; Richardson and Spirtes, 1999; Schmidt and Murphy, 2009; Claassen and Heskes, 2010; Peters et al., 2010; Triantafillou et al., 2010), building on the framework of causal Bayes networks…”
• Cyclic Causal Discovery from Continuous Equilibrium Data (2013, unknown): “An alternative approach to dealing with cycles was proposed by Schmidt and Murphy (2009).”
• Learning linear cyclic causal models with latent variables (2012, background): “In these cases a cyclic model provides the natural representation, and one needs to make use of causal discovery procedures that do not rely on acyclicity (Richardson, 1996; Schmidt and Murphy, 2009; Itani et al., 2008).”
• On Causal Discovery with Cyclic Additive Noise Models (2011, background): “Researchers have also studied cyclic causal models with discrete variables [9, 10].”