Finite-time convergence policies in state-dependent social networks.
Daniel SilvestrePaulo Andre Nobre RosaJoão Pedro HespanhaCarlos SilvestrePublished in: ACC (2015)
Keyphrases
- state dependent
- optimal policy
- social networks
- markov decision processes
- optimal production
- steady state
- finite state
- reinforcement learning
- social network analysis
- queueing networks
- dynamic programming
- continuous state
- state space
- long run
- infinite horizon
- sufficient conditions
- finite horizon
- service rates
- product form
- control policies
- asymptotically optimal
- complex networks
- multistage
- finite number
- stationary distribution
- customer demand
- average cost
- arrival rate
- markov chain
- initial state
- single server
- network structure