Generative Planning for Temporally Coordinated Exploration in Reinforcement Learning.
Haichao ZhangWei XuHaonan YuPublished in: ICLR (2022)
Keyphrases
- reinforcement learning
- action selection
- exploration strategy
- active exploration
- spatio temporal
- multiple robots
- state space
- multi agent
- generative model
- temporal difference
- planning problems
- exploration exploitation
- deterministic domains
- macro actions
- machine learning
- cooperative
- autonomous learning
- optimal policy
- temporal information
- domain independent
- heuristic search
- blocks world
- model based reinforcement learning
- unsupervised learning
- data driven
- complex domains
- partially observable
- reinforcement learning algorithms
- model free
- search space
- active learning
- reward shaping
- exploration exploitation tradeoff
- partial observability
- stochastic domains
- reinforcement learning problems
- function approximation
- markov decision problems
- transfer learning
- discriminative learning
- motion planning