Channel Selection for Network-assisted D2D Communication via No-Regret Bandit Learning with Calibrated Forecasting.
Setareh MaghsudiSlawomir StanczakPublished in: CoRR (2014)
Keyphrases
- online learning
- reinforcement learning
- learning process
- recurrent networks
- communication networks
- learning algorithm
- connectionist networks
- communication links
- radial basis function network
- communication channels
- communication systems
- computer networks
- short term
- communication cost
- multi channel
- communication overhead
- social network analysis
- bandit problems
- pose estimation