Robust Q-learning
From MaRDI portal
Abstract: Q-learning is a regression-based approach that is widely used to formalize the development of an optimal dynamic treatment strategy. Finite dimensional working models are typically used to estimate certain nuisance parameters, and misspecification of these working models can result in residual confounding and/or efficiency loss. We propose a robust Q-learning approach which allows estimating such nuisance parameters using data-adaptive techniques. We study the asymptotic behavior of our estimators and provide simulation studies that highlight the need for and usefulness of the proposed method in practice. We use the data from the "Extending Treatment Effectiveness of Naltrexone" multi-stage randomized trial to illustrate our proposed methods.
Recommendations
Cites work
- \({\mathcal Q}\)-learning
- Q- and A-learning methods for estimating optimal dynamic treatment regimes
- A robust method for estimating optimal treatment regimes
- Asymptotics of cross-validated risk estimation in estimator selection and performance assess\-ment
- Bias-reduced doubly robust estimation
- Cancer clinical trials. Current and controversial issues in design and analysis
- Consistency of random forests
- Demystifying double robustness: a comparison of alternative strategies for estimating a population mean from incomplete data
- Double/debiased machine learning for treatment and structural parameters
- Doubly robust nonparametric inference on the average treatment effect
- Doubly-robust dynamic treatment regimen estimation via weighted least squares
- Doubly-robust estimators of treatment-specific survival distributions in observational studies with stratified sampling
- Dynamic treatment regimes: technical challenges and applications
- Estimating Exposure Effects by Modelling the Expectation of Exposure Conditional on Confounders
- Estimating individualized treatment rules using outcome weighted learning
- High-dimensional \(A\)-learning for optimal dynamic treatment regimes
- Improving efficiency and robustness of the doubly robust estimator for a population mean with incomplete data
- Incorporating patient preferences into estimation of optimal individualized treatment rules
- Inference for optimal dynamic treatment regimes using an adaptive m-out-of-n bootstrap scheme
- New statistical learning methods for estimating optimal dynamic treatment regimes
- Optimal Dynamic Treatment Regimes
- Oracle inequalities for multi-fold cross validation
- Reinforcement learning strategies for clinical trials in nonsmall cell lung cancer
- Robust estimation of optimal dynamic treatment regimes for sequential treatment decisions
- Root-N-Consistent Semiparametric Regression
- Semiparametric Regression for Repeated Outcomes with Nonignorable Nonresponse
- Semiparametric theory and missing data.
- Statistical methods for dynamic treatment regimes. Reinforcement learning, causal inference, and personalized medicine
- Super Learner
- Unified methods for censored longitudinal data and causality
- Using the Standardized Difference to Compare the Prevalence of a Binary Variable Between Two Groups in Observational Research
- Valid post-selection inference
Cited in
(23)- Generalization error bounds of dynamic treatment regimes in penalized regression-based learning
- Q-learning for estimating optimal dynamic treatment rules from observational data
- Interactive model building for Q-learning
- Q-learning with censored data
- A cure-rate model for Q-learning: estimating an adaptive immunosuppressant treatment strategy for allogeneic hematopoietic cell transplant patients
- The QLBS Q-Learner goes NuQLear: fitted Q iteration, inverse RL, and option portfolios
- Adaptive Q-learning
- Adaptive treatment and robust control
- Rejoinder to “Reader reaction to ‘Outcome‐adaptive Lasso: Variable selection for causal inference’ by Shortreed and Ertefaie (2017)”
- Optimal Treatment Regimes: A Review and Empirical Comparison
- Flexible inference of optimal individualized treatment strategy in covariate adjusted randomization with multiple covariates
- Deep spectral Q-learning with application to mobile health
- Penalized robust learning for optimal treatment regimes with heterogeneous individualized treatment effects
- On ``Reflections on the concept of optimality of single decision point treatment regimes
- Relative sparsity for medical decision problems
- Double robust estimation of optimal partially adaptive treatment strategies: an application to breast cancer treatment using hormonal therapy
- Deep spatial Q-learning for infectious disease control
- Transfer Q-learning for finite-horizon Markov decision processes
- Using machine learning to improve control for confounding in the dynamic weighted ordinary least squares estimator of optimal adaptive treatment strategies
- Individualized treatment rules based on adaptive transfer-dragonnet
- Sensitivity analysis for constructing optimal regimes in the presence of treatment non-compliance and two active treatments
- Dynamic treatment regimes on dyadic networks
- Bayesian empirical likelihood regression for semiparametric estimation of optimal dynamic treatment regimes
This page was built for publication: Robust Q-learning
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5857152)
