Cost-effectively Identifying Causal Effect When Only Response Variable Observable

Part of Proceedings of the International Conference on Machine Learning 1 pre-proceedings (ICML 2020)

Bibtex »Metadata »Paper »Supplemental »

Bibtek download is not availble in the pre-proceeding


Tian-Zuo Wang, Xi-Zhu Wu, Sheng-Jun Huang, Zhi-Hua Zhou


<p>In many real tasks, we care about how to make decisions other than mere predictions on an event, e.g. how to increase the revenue next month instead of knowing it will drop. The key is to identify the causal effects on the desired event. Pearl proposed do-calculus to make it given the knowledge of causal structure (Pearl, 2009). But sometimes, we have to discover it at first. In this paper, we propose a novel solution for this challenging task where only the response variable is observable under intervention. By an active strategy introducing limited interventions and exploiting the exact distribution of the response variable, the proposed approach can cost-effectively identify the causal effect of each intervention, and thus guide the decision-making. Theoretical analysis along with empirical studies is presented to show that our approach can achieve causal effect identification with fewer interventions.</p>