Handling Sparse Rewards in Reinforcement Learning Using Model Predictive Control.
Murad DawoodNils DenglerJorge de HeuvelMaren BennewitzPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- model predictive control
- control system
- predictive control
- markov decision processes
- state space
- function approximation
- reinforcement learning algorithms
- reward function
- model free
- optimal policy
- reward shaping
- sparse representation
- data mining
- control scheme
- neural network
- optimal control
- fuzzy logic
- decision making
- genetic algorithm
- mpc algorithm