Positivity-free Policy Learning with Observational Data

Abstract

Policy learning utilizing observational data is pivotal across various domains, with the objective of learning the optimal treatment assignment policy while adhering to specific constraints such as fairness, budget, and simplicity. This study introduces a novel positivity-free (stochastic) policy learning framework designed to address the challenges posed by the impracticality of the positivity assumption in real-world scenarios. This framework leverages incremental propensity score policies to adjust propensity score values instead of assigning fixed values to treatments. We characterize these incremental propensity score policies and establish identification conditions, employing semiparametric efficiency theory to propose efficient estimators capable of achieving rapid convergence rates, even when integrated with advanced machine learning algorithms. This paper provides a thorough exploration of the theoretical guarantees associated with policy learning and validates the proposed framework's finite-sample performance through comprehensive numerical experiments, ensuring the identification of causal effects from observational data is both robust and reliable.

References in corpus (17)

• Equality of Opportunity in Supervised Learning
• Meta-learners for Estimating Heterogeneous Treatment Effects using Machine Learning
• Statistical inference for the mean outcome under a possibly non-unique optimal treatment strategy
• Fairlearn: Assessing and Improving Fairness of AI Systems
• Semiparametric doubly robust targeted double machine learning: a review
• Identifying Dynamic Sequential Plans
• Estimating Total Treatment Effect in Randomized Experiments with Unknown Network Structure
• Causal Estimation for Text Data with (Apparent) Overlap Violations
• Staggered Rollout Designs Enable Causal Inference Under Interference Without Network Knowledge
• Policy Learning under Biased Sample Selection
• Bayesian Safe Policy Learning with Chance Constrained Optimization: Application to Military Security Assessment during the Vietnam War
• Off-policy evaluation beyond overlap: partial identification through smoothness
• Average treatment effect on the treated, under lack of positivity
• Policy learning "without" overlap: Pessimism and generalized empirical Bernstein's inequality
• Robust and Agnostic Learning of Conditional Distributional Treatment Effects
• Causal Effect Estimation after Propensity Score Trimming with Continuous Treatments
• Risk Minimization from Adaptively Collected Data: Guarantees for Supervised and Policy Learning