Modeling Discrete Interventional Data using Directed Cyclic Graphical Models
This work addresses the challenge of representing and learning from interventional data in domains like biology, though it appears incremental as it builds on existing graphical model frameworks.
The paper tackles the problem of modeling discrete interventional data by proposing a representation using directed cyclic graphical models, which allows for global normalization and convex optimization for parameter estimation, and it is evaluated on simulated and flow cytometry data.
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.