Reconfigurable disruption tolerant routing via Reinforcement Learning.
Tae-Hyung KimLarry D. PyeattDonald C. WunschPublished in: IJCNN (2009)
Keyphrases
- reinforcement learning
- general purpose
- function approximation
- routing algorithm
- low cost
- routing protocol
- routing problem
- reconfigurable architecture
- reinforcement learning algorithms
- multi agent
- shortest path
- optimal policy
- markov decision processes
- model free
- interconnection networks
- machine learning
- learning problems
- transfer learning
- state space
- dynamic programming
- ad hoc networks
- digital signal
- hardware implementation
- inter domain
- temporal difference
- multi agent reinforcement learning
- function approximators
- traffic engineering
- neural network
- end to end
- policy search
- genetic algorithm
- wireless ad hoc networks
- markov decision process
- multi agent systems
- action selection
- wireless sensor networks
- optimal control
- mobile robot