Causality Learning With Wasserstein Generative Adversarial Networks. (arXiv:2206.01496v1 [cs.LG])

Conventional methods for causal structure learning from data face significant
challenges due to combinatorial search space. Recently, the problem has been
formulated into a continuous optimization framework with an acyclicity
constraint to learn Directed Acyclic Graphs (DAGs). Such a framework allows the
utilization of deep generative models for causal structure learning to better
capture the relations between data sample distributions and DAGs. However, so
far no study has experimented with the use of Wasserstein distance in the
context of causal …

arxiv causality generative adversarial networks learning networks