Hierarchical Advantage for Reinforcement Learning in Parameterized Action Space.
Zhejie HuTomoyuki KanekoPublished in: CoG (2021)
Keyphrases
- action space
- reinforcement learning
- state space
- state and action spaces
- markov decision processes
- real valued
- continuous state
- control policies
- reinforcement learning methods
- continuous state spaces
- action selection
- state action
- stochastic processes
- function approximators
- temporal difference
- reinforcement learning algorithms
- markov decision process
- function approximation
- hidden markov models
- single agent
- machine learning
- markov decision problems
- control problems
- pairwise
- model free
- infinite horizon
- optimal control
- decision problems
- heuristic search