An LSPI Based Reinforcement Learning Approach to Enable Network Cooperation in Cognitive Wireless Sensor Network.
Milos RovcaninEli De PoorterIngrid MoermanPiet DemeesterPublished in: AINA Workshops (2013)
Keyphrases
- reinforcement learning
- wireless sensor networks
- reinforcement learning algorithms
- model free
- temporal difference
- multi agent
- reinforcement learning methods
- state space
- learning algorithm
- policy iteration
- cooperative
- function approximation
- resource limitations
- machine learning
- markov decision processes
- supervised learning
- sensor nodes
- action selection
- optimal policy
- sensor networks
- state action
- learning process
- continuous state
- policy search
- multi agent systems
- energy consumption