Explicit Planning for Efficient Exploration in Reinforcement Learning.
Liangpeng ZhangKe TangXin YaoPublished in: NeurIPS (2019)
Keyphrases
- reinforcement learning
- action selection
- planning problems
- function approximation
- partial observability
- deterministic domains
- partially observable
- state space
- heuristic search
- model free
- temporal difference
- motion planning
- machine learning
- optimal control
- partially observable markov decision processes
- planning process
- plan generation
- temporal difference learning
- blocks world
- markov decision problems
- domain independent
- macro actions
- multi agent
- problems in artificial intelligence