Institute for Systems Research

Permanent URI for this communityhttp://hdl.handle.net/1903/4375

Browse

Author: search.filters.author.Fu, Michael C.Subject: search.filters.subject.Intelligent Control Systems

Search Results

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    Randomized Difference Two-Timescale Simultaneous Perturbation Stochastic Approximation Algorithms for Simulation Optimization of Hidden Markov Models
    (2000) Bhatnagar, Shalabh; Fu, Michael C.; Marcus, Steven I.; Bhatnagar, Shashank; Marcus, Steven I.; Fu, Michael C.; ISR
    We proposetwo finite difference two-timescale simultaneous perturbationstochastic approximation (SPSA)algorithmsfor simulation optimization ofhidden Markov models. Stability and convergence of both thealgorithms is proved.

    Numericalexperiments on a queueing model with high-dimensional parameter vectorsdemonstrate orders of magnitude faster convergence using thesealgorithms over related $(N+1)$-Simulation finite difference analoguesand another two-simulation finite difference algorithm that updates incycles.

  • Thumbnail Image
    Item
    Optimal Multilevel Feedback Policies for ABR Flow Control using Two Timescale SPSA
    (1999) Bhatnagar, Shalabh; Fu, Michael C.; Marcus, Steven I.; ISR
    Optimal multilevel control policies for rate based flow control in available bit rate (ABR) service in asynchronous transfer mode (ATM) networks are obtained in the presence of information and propagation delays, using a numerically efficient two timescale simultaneous perturbation stochastic approximation (SPSA) algorithm. Numerical experiments demonstrate fast convergence even in the presence of significant delays and a large number of parametrized parameter levels.
  • Thumbnail Image
    Item
    Monotone Optimal Policies for a Transient Queueing Staffing Problem
    (1997) Fu, Michael C.; Marcus, Steven I.; Wang, I-Jeng; ISR
    We consider the problem of determining the optimal policy for staffing a queueing system over multiple periods, using a model that takes into account transient queueing effects. Formulating the problem in a dynamic programming setting, we show that the optimal policy follows a monotone optimal control by establishing the submodularity of the objective function with respect to the staffing level and initial queue size in a period. In particular, this requires proving that the system occupancy in a G/M/s queue is submodular in the number of servers and initial system occupancy.