Relational Q-Functionals: Multi-Agent Learning to Recover from Unforeseen Robot Malfunctions in Continuous Action Domains.
Yasin FindikPaul RobinetteKshitij JerathReza AzadehPublished in: CoRR (2024)
Keyphrases
- multi agent learning
- complex domains
- continuous action
- game theory
- mobile robot
- multi agent
- policy search
- artificial intelligence
- continuous state
- robot navigation
- path planning
- robotic systems
- autonomous robots
- cooperative
- resource allocation
- reinforcement learning
- machine learning
- transfer learning
- dynamic programming