Properties of the Gittins Index with Application to Optimal Scheduling
نویسندگان
چکیده
We consider the optimal scheduling problem for a single-server queue without arrivals. We allow preemptions, and our purpose is to minimize the expected flow time. The optimal non-anticipating discipline is known to be the Gittins index policy, which, however, is defined in an implicit way. Until now, its general behaviour in this specific problem has been characterized only in a few special cases. In this paper, we give as complete a characterization as possible. It turns out that the optimal policy always belongs to the family of Multi-Level-Processor-Sharing (MLPS) disciplines.
منابع مشابه
On the optimality of the Gittins index rule for multi-armed bandits with multiple plays
We investigate the general multi-armed bandit problem with multiple servers. We determine a condition on the reward processes su1⁄2cient to guarantee the optimality of the strategy that operates at each instant of time the projects with the highest Gittins indices. We call this strategy the Gittins index rule for multi-armed bandits with multiple plays, or brie ̄y the Gittins index rule. We show...
متن کاملFinite dimensional algorithms for the hidden Markov model multi-armed bandit problem
The multi-arm bandit problem is widely used in scheduling of traffic in broadband networks, manufacturing systems and robotics. This paper presents a finite dimensional optimal solution to the multi-arm bandit problem for Hidden Markov Models. The key to solving any multi-arm bandit problem is to compute the Gittins index. In this paper a finite dimensional algorithm is presented which exactly ...
متن کاملOptimal Sensor Scheduling via Non-standard Multi-armed Bandit Formulation
This paper addresses optimal sensor scheduling for maximizing network lifetime. We formulate this problem as a nonstandard multi-armed bandit process with non-discounted reward and finite stopping time. We find that the optimal strategy should choose the sensor that has the largest Gittins’ index. Exploiting the underlying structure of sensor scheduling problem, we reduce the computational comp...
متن کاملSome results on the Gittins index for a normal reward process
Abstract: We consider the Gittins index for a normal distribution with unknown mean θ and known variance where θ has a normal prior. In addition to presenting some monotonicity properties of the Gittins index, we derive an approximation to the Gittins index by embedding the (discrete-time) normal setting into the continuous-time Wiener process setting in which the Gittins index is determined by...
متن کاملHidden Markov model multiarm bandits: a methodology for beam scheduling in multitarget tracking
In this paper, we derive optimal and suboptimal beam scheduling algorithms for electronically scanned array tracking systems. We formulate the scheduling problem as a multiarm bandit problem involving hidden Markov models (HMMs). A finite-dimensional optimal solution to this multiarm bandit problem is presented. The key to solving any multiarm bandit problem is to compute the Gittins index. We ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2010