Hyper-Meta Reinforcement Learning with Sparse Reward.
Yun HuaXiangfeng WangBo JinWenhao LiJunchi YanXiaofeng HeHongyuan ZhaPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- function approximation
- state space
- model free
- reinforcement learning algorithms
- sparse data
- reward function
- temporal difference
- markov decision processes
- supervised learning
- eligibility traces
- sparse coding
- partially observable environments
- multi agent
- sparse representation
- high dimensional
- learning problems
- learning algorithm
- compressive sensing
- compressed sensing
- multi agent reinforcement learning
- total reward
- action selection
- transfer learning
- optimal control
- optimal policy
- partially observable
- control policy
- state action
- reinforcement learning methods
- learning process