Online Safety Property Collection and Refinement for Safe Deep Reinforcement Learning in Mapless Navigation.
Luca MarzariEnrico MarchesiniAlessandro FarinelliPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- online learning
- real time
- state space
- database
- robotic control
- balancing exploration and exploitation
- information space
- function approximation
- markov decision processes
- document collections
- website
- learning algorithm
- multi agent
- optimal policy
- search engine
- model free
- reinforcement learning algorithms
- temporal difference learning
- reinforcement learning methods
- multi agent reinforcement learning
- machine learning
- neural network