On the Challenges of using Reinforcement Learning in Precision Drug Dosing: Delay and Prolongedness of Action Effects.
Sumana BasuMarc-André LegaultAdriana Romero-SorianoDoina PrecupPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- action selection
- indirect effects
- partially observable domains
- action space
- function approximation
- key issues
- lessons learned
- high precision
- reward shaping
- technical challenges
- precision and recall
- real world
- open issues
- reasoning about actions
- transition model
- information retrieval
- data mining
- agent learns
- machine learning
- policy search
- learning algorithm
- average precision
- multi agent
- human actions
- supervised learning