An Application of Inverse Reinforcement Learning to Medical Records of Diabetes Treatment

By Hideki Asoh, Masanori Shiro, Shotaro Akaho, Toshihiro Kamishima. 2013

24 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Reinforcement Learning in Healthcare: A Survey (2019, methodology): “The variance during the policy and reward learning can amplify the bias in each learning step, potentially leading to divergent solutions that can be of little use in practice [11].”
Maximum Entropy Deep Inverse Reinforcement Learning (2015, background): “The network serves to model the relationship between input features and final reward map. as a tool to better understand human decisions or medicine (Asoh et al., 2013) to predict patient response to treatment.”
Bayesian Robust Optimization for Imitation Learning (2020, background): “While this makes sense for highly risk-sensitive domains such as autonomous driving [48, 58] or medicine [5, 28], in other domains such as a robot vacuuming office carpets, we may also be interested in efficiency and performance, rather than pure risk-aversion.”
Optimal medication dosing from suboptimal clinical examples: A deep reinforcement learning approach (2016, methodology): “Application of RL to design of clinical trials and adjustment of clinical treatments have been previously suggested in the literature [15], [16], [17].”
The Actor Search Tree Critic (ASTC) for Off-Policy POMDP Learning in Medical Decision Making (2018, background): “For example, [9] proposed hemoglobin-A1c dosages for diabetes patients by implementing Markov chain Monte Carlo sampling to infer posterior distribution over rewards given observed states and actions.”
Representation and Reinforcement Learning for Personalized Glycemic Control in Septic Patients (2017, methodology): “Related to glycemic control, some studies utilize RL to design clinical trials and adjust clinical treatments [Asoh et al., 2013, Bothe et al., 2014].”