Events

Dr. Huaiyang Zhong
Postdoctoral Fellow
Harvard Medical School/Massachusetts General Hospital
Harvard University

Healthcare decision-making is interesting and challenging because of the probabilistic nature of many healthcare decision problems and because of the range of decision-makers involved. In this talk, I will present one of my studies that comes from a real-life medical decision-making problem and eventually helps address an important theoretical problem in risk-sensitive sequential decision making. Specifically, unlike the traditional Markov decision processes that maximize the expected value of the cumulative reward, a decision-maker might want to optimize the quantile value of the cumulative reward depending on their risk preferences. In the solution, we provide analytical results characterizing the optimal QMDP value function and present a dynamic programming-based algorithm to solve for the optimal policy. The algorithm also extends to the MDP problem with a conditional value-at-risk objective. We illustrate the practical relevance of our model by evaluating it on an HIV treatment initiation problem, in which patients aim to balance the potential benefits and risks of the treatment.