Recently there have been many efforts to imbue deep-learning models with the ability to perform causal inference. This has been motivated primarily by the inability of traditional correlative models to make predictions on interventional and counterfactual questions, as well as the "explainability" of causal graphical models. In a recent paper I applied a special type of deep-learning objective function called a "moment-matching" loss function to the problem of causal inference. Check out the GitHub page for a top-to-bottom demo!