Accelerating Lifelong Reinforcement Learning via Reshaping Rewards.
Kun ChuXianchao ZhuWilliam ZhuPublished in: SMC (2021)
Keyphrases
- reinforcement learning
- function approximation
- markov decision processes
- reinforcement learning algorithms
- reward function
- temporal difference
- state space
- learning algorithm
- competence development
- optimal policy
- learning problems
- optimal control
- model free
- lifelong learning
- learning activities
- reinforcement learning methods
- robotic control
- least squares
- reward shaping
- mobile robot
- partially observable
- neural network
- total reward
- hidden state
- stochastic approximation
- policy iteration
- social networks
- dynamic programming
- action selection
- learning processes
- transfer learning