Active Learning of Markov Decision Processes using Baum-Welch algorithm.
Giovanni BacciAnna IngólfsdóttirKim G. LarsenRaphaël ReynouardPublished in: ICMLA (2021)
Keyphrases
- markov decision processes
- dynamic programming
- learning algorithm
- active learning
- objective function
- optimal solution
- computational complexity
- state space
- np hard
- hidden markov models
- average reward
- k means
- model based reinforcement learning
- search space
- factored mdps
- neural network
- evolutionary algorithm
- pairwise
- expectation maximization
- optimal policy
- reinforcement learning
- convergence rate
- genetic algorithm
- action sets