Modular Lifelong Reinforcement Learning via Neural Composition.
Jorge A. MendezHarm van SeijenEric EatonPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- fitted q iteration
- network architecture
- neural network
- function approximation
- state space
- optimal policy
- st century
- competence development
- machine learning
- temporal difference
- model free
- nonlinear predictive control
- web service composition
- sensory inputs
- bio inspired
- robotic control
- neural computation
- function approximators
- reinforcement learning algorithms
- supervised learning
- temporal difference learning
- reinforcement learning methods
- neural model
- hebbian learning
- neural fuzzy
- modular neural networks
- learning algorithm
- genetic algorithm
- technology enhanced learning