Towards Transparency of TD-RL Robotic Systems with a Human Teacher.
Marco MatareseSilvia RossiAlessandra SciuttiFrancesco ReaPublished in: CoRR (2020)
Keyphrases
- imitation learning
- human teacher
- reinforcement learning
- robotic systems
- temporal difference
- reinforcement learning algorithms
- td learning
- function approximation
- model free
- state space
- reinforcement learning methods
- markov decision processes
- vision system
- learning process
- evaluation function
- mobile robot
- learning algorithm
- action selection
- policy iteration
- machine learning
- multi agent
- learning mechanism
- control problems
- dynamic programming
- function approximators
- learning problems
- multi modal
- video sequences
- optimal policy
- reward function
- active learning
- control strategies
- optimal control
- step size