Causal Learner: A Toolbox for Causal Structure and Markov Blanket Learning

TL;DR

Causal Learner addresses the need for an accessible, feature-rich toolbox for causal structure and Markov blanket learning. It combines data-generation in R, a broad MATLAB-based algorithm suite, and comprehensive evaluation metrics to enable end-to-end experimentation. The toolbox extends prior work by offering more algorithms, explicit data-generation capabilities, and open-source accessibility, facilitating research and practical deployment. Its open architecture and documented usage support rapid development and evaluation of new causal learning methods with real and simulated data.

Abstract

Causal Learner is a toolbox for learning causal structure and Markov blanket (MB) from data. It integrates functions for generating simulated Bayesian network data, a set of state-of-the-art global causal structure learning algorithms, a set of state-of-the-art local causal structure learning algorithms, a set of state-of-the-art MB learning algorithms, and functions for evaluating algorithms. The data generation part of Causal Learner is written in R, and the rest of Causal Learner is written in MATLAB. Causal Learner aims to provide researchers and practitioners with an open-source platform for causal learning from data and for the development and evaluation of new causal learning algorithms. The Causal Learner project is available at http://bigdata.ahu.edu.cn/causal-learner.

Figures (3)

• Figure 1: Examples of a global causal structure, local causal structure, and Markov blanket ($T$ in black is a target node).
• Figure 2: The architecture of Causal Learner.
• Figure 3: An example of using Causal Learner to learn a global causal structure.