Inference on the Value of Policy Improvement in Adaptive Experiments

stat.ML, cs.LG, econ.EM, math.ST, stat.TH

Submitted: 2026-10-11

Updated: 2026-10-11

Comments: 97 pages, 10 figures, 13 tables

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Sources

• Sufficient and insufficient conditions for the stochastic convergence of Ces\`{a}ro means
• Inference on Optimal Dynamic Policies via Softmax Approximation
• Doubly Robust Policy Evaluation and Learning
• Sample-efficient Nonstationary Policy Evaluation for Contextual Bandits
• Fast Best-in-Class Regret for Contextual Bandits
• Cramming Contextual Bandits for On-policy Statistical Evaluation
• On the Lower Confidence Band for the Optimal Welfare in Policy Learning
• Inference on Optimal Policy Values and Other Irregular Functionals via Softmax Smoothing

Related papers

• Behavior of prediction performance metrics with rare events
• Optimal Estimation of Generic Dynamics by Path-Dependent Neural Jump ODEs
• A Posterior-Dynamics Framework for Imaging Inverse Problems with Pretrained Diffusion Priors
• Scalable Posterior Uncertainty for Flexible Density-Based Clustering
• Efficient quadratic entropy with distance sketches
• Embedding-Bias in Conditional Independence Testing