PubMed · 2012 · 71 citations · 41 references
BusinessDecision SciencesPsychiatryClinical Decision SupportEvidence-based PracticeMarkov Decision ProcessesClinical Decision-makingDecision-making ModelsReinforcement LearningReinforcement Learning FoundationsValue FunctionsValue IterationClinical Decision Support ModelsDecision Aid DesignDecision Aid DevelopmentDecision-making AlgorithmsFinite-horizon Value IterationLinear ApproximationLinear Function ApproximationSequential Decision Support
We present a general and detailed development of an algorithm for finite-horizon fitted-Q iteration with an arbitrary number of reward signals and linear value function approximation using an arbitrary number of state features. This includes a detailed treatment of the 3-reward function case using triangulation primitives from computational geometry and a method for identifying globally dominated actions. We also present an example of how our methods can be used to construct a real-world decision aid by considering symptom reduction, weight gain, and quality of life in sequential treatments for schizophrenia. Finally, we discuss future directions in which to take this work that will further enable our methods to make a positive impact on the field of evidence-based clinical decision support.
41
Reinforcement Learning: An Introduction
IEEE Transactions on Neural Networks · 2005 · 25.8K citations
Evidence based medicine: what it is and what it isn't
David L. Sackett, William Rosenberg, J. A. Muir Gray et al. · BMJ · 1996 · 15.7K citations · Full text