Network Interference
CLT under Local Dependence: When Observations Are Connected
A dependency-graph CLT, its Stein-method proof, and Berry–Esseen bounds under local dependence.
Blog
Dynamic Treatment Regimes
Direct Learning of DTRs, Part iii: Multiple Treatments and SDSS
Pairwise score surrogates, simultaneous optimization, and a multistage Fisher-consistency proof.
Dynamic Treatment Regimes
Direct Learning of DTRs, Part ii: Smooth Surrogates and DTRESLO
From one-stage outcome-weighted learning to a smooth two-stage objective, with a value-regret proof.
Dynamic Treatment Regimes
Direct Learning of DTRs, Part i: Doubly Robust Value Search
Augmented value estimation, a double-robustness proof, and joint search over treatment rules.
Offline Policy Learning
Offline Policy Learning from Logged Data
Q-learning and backward induction for treatment decisions using logged longitudinal data.
Surrogate Losses & Consistency
Surrogate Loss Functions and Fisher Consistency in Binary Classification
Surrogate risk minimization and Fisher consistency for binary classification.
Coming soon
Offline Policy Learning
Support-Aware Offline Policy Selection
Offline policy selection when logged data provide limited coverage of candidate actions.
Coming soon
Network Interference
Policy Learning under Network Interference
Single-stage policy learning when treatments affect outcomes across a network.