Deep Reinforcement Learning for Autonomous Model-Free Navigation with Partial Observability.
Daniel TapiaJuan ParrasSantiago ZazoPublished in: EUSIPCO (2019)
Keyphrases
- model free
- partial observability
- reinforcement learning
- partially observable
- reinforcement learning algorithms
- function approximation
- state space
- temporal difference
- markov decision process
- policy iteration
- learning capabilities
- learning agent
- partially observable markov decision processes
- supervised learning
- markov decision processes
- planning under partial observability
- robotic systems
- transfer learning
- optimal policy
- function approximators
- multi agent
- learning algorithm
- policy evaluation
- machine learning
- belief state
- learning process