Action-Bounding for Reinforcement Learning in Energy Harvesting Communication Systems.
Heasung KimHeecheol YangYeongmo KimJungwoo LeePublished in: GLOBECOM (2018)
Keyphrases
- communication systems
- reinforcement learning
- action selection
- information processing systems
- action space
- partially observable domains
- reward shaping
- underwater acoustic
- computer systems
- channel estimation
- blind equalization
- upper bound
- state action
- energy consumption
- communication technologies
- multiple access
- function approximation
- transition model
- ultra wideband
- state space
- temporal difference
- model free
- multi agent
- agent learns
- learning algorithm
- markov decision process
- reinforcement learning algorithms
- markov decision processes
- user interface
- optimal policy
- policy search
- database systems
- databases