Score-Aware Policy-Gradient Methods and Performance Guarantees using Local Lyapunov Conditions: Applications to Product-Form Stochastic Networks and Queueing Systems.
Céline ComteMatthieu JonckheereJaron SandersAlbert Senen-CerdaPublished in: CoRR (2023)
Keyphrases
- lower bound
- queueing systems
- queueing networks
- product form
- state dependent
- steady state
- arrival rate
- queue length
- sufficient conditions
- stationary distribution
- markov processes
- heavy traffic
- single server
- long run
- markov chain
- policy gradient methods
- queueing model
- adaptive control
- control problems
- service times
- control law
- multi agent
- dynamical systems