RLProph: a dynamic programming based reinforcement learning approach for optimal routing in opportunistic IoT networks.
Deepak Kumar SharmaJoel J. P. C. RodriguesVidushi VashishthAnirudh KhannaAnshuman ChhabraPublished in: Wirel. Networks (2020)
Keyphrases
- dynamic programming
- reinforcement learning
- optimal control
- markov decision processes
- optimal policy
- state space
- network topologies
- function approximation
- learning algorithm
- network structure
- network topology
- wireless ad hoc networks
- locally optimal
- linear programming
- cloud computing
- management system
- policy search
- network design
- approximate dynamic programming
- switched networks
- path selection
- network nodes
- machine learning
- temporal difference
- multi agent
- complex networks