Distilling a Hierarchical Policy for Planning and Control via Representation and Reinforcement Learning.
Jung-Su HaYoung-Jin ParkHyeok-Joo ChaeSoon-Seo ParkHan-Lim ChoiPublished in: ICRA (2021)
Keyphrases
- reinforcement learning
- action selection
- optimal policy
- control policy
- control problems
- control policies
- partially observable
- hierarchical reinforcement learning
- reinforcement learning problems
- hierarchical decomposition
- markov decision processes
- optimal control
- hierarchical representation
- control system
- partially observable markov decision processes
- function approximation
- dynamic programming
- markov decision process
- action space
- markov decision problems
- control rules
- state space
- machine learning
- continuous state
- partially observable domains
- learning algorithm
- planning problems
- policy iteration
- control method
- partial observability
- heuristic search
- robotic control
- domain independent
- control strategy
- policy search
- policy gradient
- adaptive control
- control strategies
- temporal difference
- inverted pendulum
- reinforcement learning algorithms
- complex domains
- robot control