Identifying Distinct, Effective Treatments for Acute Hypotension with SODA-RL: Safely Optimized Diverse Accurate Reinforcement Learning.
Joseph FutomaMuhammad A. MasoodFinale Doshi-VelezPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- function approximation
- high quality
- computationally efficient
- markov decision processes
- model free
- multi agent
- reinforcement learning algorithms
- wide variety
- control problems
- learning process
- high accuracy
- state space
- learning algorithm
- temporal difference learning
- action selection
- supervised learning
- action space
- markov decision process
- learning capabilities
- multi agent reinforcement learning
- autonomous learning
- reinforcement learning methods
- early detection
- dynamic programming
- highly accurate
- decision support system
- optimal policy
- image restoration