Neurorobotic reinforcement learning for domains with parametrical uncertainty.
Camilo AmayaAxel von ArnimPublished in: Frontiers Neurorobotics (2023)
Keyphrases
- reinforcement learning
- partial observability
- transfer learning
- partially observable domains
- real world
- inherent uncertainty
- function approximation
- deterministic domains
- markov decision processes
- machine learning
- incomplete information
- action space
- possibility theory
- model free
- temporal difference
- reinforcement learning algorithms
- neural network
- dynamic programming
- robotic control
- genetic algorithm
- partially observable
- learning algorithm
- uncertain data
- conditional probabilities
- decision problems
- learning process
- least squares
- state space