Hierarchical Average-Reward Linearly-solvable Markov Decision Processes.
Guillermo InfanteAnders JonssonVicenç GómezPublished in: CoRR (2024)
Keyphrases
- markov decision processes
- average reward
- hierarchical reinforcement learning
- optimal policy
- policy iteration
- discounted reward
- stochastic games
- semi markov decision processes
- finite state
- long run
- dynamic programming
- reinforcement learning
- state space
- optimality criterion
- total reward
- reinforcement learning algorithms
- state and action spaces
- np hard
- factored mdps
- decision theoretic planning
- planning under uncertainty
- infinite horizon
- markov decision process
- average cost
- partially observable
- actor critic
- decision problems
- computational complexity
- discount factor
- policy gradient
- reward function
- model free
- markov chain
- optimal solution