Risk-Sensitive Reinforcement Learning for URLLC Traffic in Wireless Networks.
Nesrine Ben KhalifaMohamad AssaadMérouane DebbahPublished in: CoRR (2018)
Keyphrases
- wireless networks
- risk sensitive
- reinforcement learning
- model free
- optimal control
- markov decision processes
- medium access control
- markov decision problems
- control policies
- wireless communication
- function approximation
- reinforcement learning algorithms
- intelligent transportation systems
- packet scheduling
- state space
- utility function
- optimal policy
- ad hoc networks
- network traffic
- traffic flow
- dynamic programming
- cellular networks
- mobile networks
- multi hop
- temporal difference
- policy iteration
- real time
- machine learning
- multi agent
- network coding
- multimedia services
- low bandwidth
- partially observable
- action space
- average reward
- average cost
- control policy
- wireless mesh networks
- control strategies
- cross layer
- infinite horizon
- markov chain
- optimality criterion
- computational complexity
- expected utility