policytree: Policy learning via doubly robust empirical welfare maximization over trees
Erik Sverdrup, Ayush Kanodia, Zhengyuan Zhou, Susan Athey, Stefan Wager
Abstract
The problem of learning treatment assignment policies from randomized or observational data arises in many fields. For example, in personalized medicine, we seek to map patient observables (like age, gender, heart pressure, etc.) to a treatment choice using a data-driven rule.
Topics & Concepts
MaximizationWelfarePolicy learningComputer scienceEconomicsPublic economicsMicroeconomicsEconometricsArtificial intelligenceMachine learningMarket economyAdvanced Causal Inference TechniquesWater resources management and optimizationGame Theory and Voting Systems