Exploring Safer Behaviors for Deep Reinforcement Learning.
Enrico MarchesiniDavide CorsiAlessandro FarinelliPublished in: AAAI (2022)
Keyphrases
- reinforcement learning
- real robot
- function approximation
- state space
- optimal control
- multi agent
- markov decision processes
- machine learning
- partially observable
- reinforcement learning algorithms
- model free
- human behavior
- transfer learning
- temporal difference
- learning algorithm
- sensory inputs
- multi agent reinforcement learning
- multi robot
- finite state machines
- mobile robot
- dynamic programming
- data mining
- deep learning
- function approximators
- behavior recognition