Knowledge Transfer for Learning Markov Equivalence Classes
Proceedings of the 10th International Conference on Probabilistic Graphical Models, PMLR 138:377-388, 2020.
There are domains, such as in biology, medicine, and neuroscience, where the causal relations vary across members of a population, and where it may be difficult to collect data for some specific members. For these domains, it is convenient to develop algorithms that, from small sample sizes, can discover the specific causal relations of a subject. Learning these subject-specific models with the existing causal discovery algorithms could be difficult. Most of them were designed to find the common causal relations of a population in the large sample limit. Although transfer learning techniques have shown to be useful for improving predictive associative models learned with limited data sets, their application in the field of causal discovery has not been sufficiently explored. In this paper, we propose a knowledge transfer algorithm for discovering Markov equivalence classes for subject-specific causal models. We explore transferring weighted instances of auxiliary data sets, according to their relevance, for improving models learned with limited sample sizes. Experimental results on data sets generated from simulated and benchmark causal Bayesian networks show that our method outperforms in adjacency and arrowhead recovery the base and a similar knowledge transfer discovery methods.