Handling Sparse Rewards in Reinforcement Learning Using Model Predictive Control.
Murad DawoodNils DenglerJorge de HeuvelMaren BennewitzPublished in: ICRA (2023)
Keyphrases
- reinforcement learning
- model predictive control
- control system
- predictive control
- markov decision processes
- function approximation
- reinforcement learning algorithms
- reward function
- model free
- machine learning
- learning algorithm
- state space
- optimal policy
- reward shaping
- sparse representation
- decision makers
- optimal control
- control scheme
- action selection
- fuzzy logic
- control policy
- decision making
- neural network
- supervised learning
- least squares
- dynamic programming