Scoring-Aggregating-Planning: Learning task-agnostic priors from interactions and sparse rewards for zero-shot generalization.
Huazhe XuBoyuan ChenYang GaoTrevor DarrellPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- learning process
- efficient learning
- object recognition
- prior knowledge
- mobile learning
- unsupervised learning
- learning tasks
- learning systems
- knowledge acquisition
- supervised learning
- learning algorithm
- active learning
- high dimensional
- multi task
- explanation based learning
- learning environment
- macro actions
- credit assignment
- search control rules