Frontiers of Counterfactual Outcome Estimation in Time Series (Invited Talk by Yan Liu)

Recent development in deep learning has spurred research advances in time series modeling and analysis. In particular, estimation of temporal counterfactual outcomes from observed history is crucial for decision-making in many domains such as healthcare and e-commerce. In this talk, I will discuss our recent work in counterfactual outcome estimation in time series, including an examination of balancing strategy for counterfactual estimation as well as a self-supervised learning framework.