PARRoT: Predictive Ad-hoc Routing Fueled by Reinforcement Learning and Trajectory Knowledge.
Benjamin SliwaCedrik SchülerManuel PatchouChristian WietfeldPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- domain knowledge
- knowledge base
- knowledge acquisition
- data mining models
- domain experts
- higher level
- knowledge based systems
- data mining techniques
- knowledge representation
- expert systems
- multi agent
- state space
- knowledge management
- prior knowledge
- learning process
- design process
- markov decision processes
- knowledge extraction
- predictive model
- previously learned
- data sets