Safe Offline Reinforcement Learning Through Hierarchical Policies.
Shaofan LiuShiliang SunPublished in: PAKDD (2) (2022)
Keyphrases
- reinforcement learning
- optimal policy
- hierarchical reinforcement learning
- policy search
- markov decision process
- semi markov decision process
- fitted q iteration
- control policies
- reward function
- function approximation
- reinforcement learning agents
- model free
- learning process
- state space
- reinforcement learning algorithms
- partially observable markov decision processes
- state abstraction
- markov decision problems
- control policy
- continuous state
- policy gradient methods
- multiscale
- hierarchical clustering
- hierarchical structure
- multi agent
- neural network
- average reward
- decision processes
- action space
- temporal difference
- multi agent reinforcement learning
- markov decision processes
- transfer learning
- state and action spaces
- total reward
- dynamic programming
- cooperative multi agent systems