Graph Neural Networks for Relational Inductive Bias in Vision-based Deep Reinforcement Learning of Robot Control.
Marco OlivaSoubarna BanikJosip JosifovskiAlois C. KnollPublished in: CoRR (2022)
Keyphrases
- robot control
- neural network
- reinforcement learning
- inductive bias
- mobile robot
- machine learning
- autonomous robots
- supervised learning
- pac learning
- inductive learning
- real time
- learning algorithm
- function approximation
- function approximators
- artificial neural networks
- domain knowledge
- vision system
- neural network model
- hypothesis space
- back propagation
- domain theory
- active learning
- genetic algorithm
- previously learned
- fuzzy systems
- reinforcement learning algorithms
- semi supervised learning
- fuzzy logic
- decision trees