Investigation of reward functions for controlling blood glucose level using reinforcement learning.
Lehel Dénes-FazakasMáté SiketLászló SzilágyiGyörgy EignerLevente KovácsPublished in: SACI (2023)
Keyphrases
- reinforcement learning
- reward function
- markov decision processes
- policy search
- reinforcement learning algorithms
- optimal policy
- state space
- markov decision process
- blood glucose
- partially observable
- social networks
- blood pressure
- diabetic patients
- state variables
- transition probabilities
- multiple agents
- probabilistic model
- function approximation
- function approximators
- state action
- markov decision problems
- continuous state
- dynamic programming
- inverse reinforcement learning