Skill Reward for Safe Deep Reinforcement Learning.
Jiangchang ChengFumin YuHongliang ZhangYinglong DaiPublished in: UbiSec (2021)
Keyphrases
- reinforcement learning
- eligibility traces
- function approximation
- state space
- multi agent
- behavioural cloning
- learning algorithm
- reinforcement learning algorithms
- model free
- control strategies
- reward function
- markov decision processes
- transfer learning
- supervised learning
- total reward
- learning process
- learning problems
- optimal policy
- temporal difference
- partially observable environments
- reward shaping
- reinforcement learning methods
- learning environment
- robotic control
- policy gradient
- average reward
- learning agent
- deep learning
- partially observable