Learning Decision Rules with Observational Data

• Activity Number: 398 - Beyond Traditional Approaches: Evolving Artificial Intelligence and Machine Learning to Advance Clinical Research and Drug Development
• Type: Topic Contributed
• Date/Time: Wednesday, August 5, 2020 : 1:00 PM to 2:50 PM
• Sponsor: Biometrics Section
• Author(s): Xinkun Nie* and Stefan Wager
• Companies: Stanford University and Stanford University

Abstract

In clinical research, practitioners seek to use observational data to learn personalized treatment assignment and timing policies that would maximize patients' health outcomes. In this talk, I will present recent advances on heterogeneous treatment effect estimation and sequential decision making with observational data. In particular, I'll discuss how powerful machine learning tools can be leveraged for these tasks.