Back to Search Start Over

Learning Optimal Policies from Observational Data

Authors :
Atan, Onur
Zame, William R.
van der Schaar, M
Publication Year :
2018

Abstract

Choosing optimal (or at least better) policies is an important problem in domains from medicine to education to finance and many others. One approach to this problem is through controlled experiments/trials - but controlled experiments are expensive. Hence it is important to choose the best policies on the basis of observational data. This presents two difficult challenges: (i) missing counterfactuals, and (ii) selection bias. This paper presents theoretical bounds on estimation errors of counterfactuals from observational data by making connections to domain adaptation theory. It also presents a principled way of choosing optimal policies using domain adversarial neural networks. We illustrate the effectiveness of domain adversarial training together with various features of our algorithm on a semi-synthetic breast cancer dataset and a supervised UCI dataset (Statlog).

Details

Database :
arXiv
Publication Type :
Report
Accession number :
edsarx.1802.08679
Document Type :
Working Paper