The (Surprising) Rate Optimality of Greedy Procedures for Large-Scale Ranking and Selection

by   Zaile Li, et al.

Large-scale ranking and selection (R S), which aims to select the best alternative with the largest mean performance from a finite set of alternatives, has emerged as an important research topic in simulation optimization. Ideal large-scale R S procedures should be rate optimal, i.e., the total sample size required to deliver an asymptotically non-zero probability of correct selection (PCS) grows at the minimal rate (linear rate) in the number of alternatives. Surprisingly, we discover that the naïve greedy procedure that keeps sampling the alternative with the largest running average performs strikingly well and appears rate optimal. To understand this discovery, we develop a new boundary-crossing perspective and prove that the greedy procedure is indeed rate optimal. We further show that the derived PCS lower bound is asymptotically tight for the slippage configuration of means with a common variance. Moreover, we propose the explore-first greedy (EFG) procedure and its enhanced version (^+ procedure) by adding an exploration phase to the naïve greedy procedure. Both procedures are proven to be rate optimal and consistent. Last, we conduct extensive numerical experiments to empirically understand the performance of our greedy procedures in solving large-scale R S problems.


page 1

page 2

page 3

page 4


Asymptotically Optimal Sampling Policy for Selecting Top-m Alternatives

We consider selecting the top-m alternatives from a finite number of alt...

Asymptotic Optimality of Myopic Ranking and Selection Procedures

Ranking and selection (R S) is a popular model for studying discrete-e...

Distributionally Robust Selection of the Best

Specifying a proper input distribution is often a challenging task in si...

Lookahead and Hybrid Sample Allocation Procedures for Multiple Attribute Selection Decisions

Attributes provide critical information about the alternatives that a de...

Selection of the Most Probable Best

We consider an expected-value ranking and selection problem where all k ...

Efficient Fully Sequential Indifference-Zone Procedures Using Properties of Multidimensional Brownian Motion Exiting a Sphere

We consider a ranking and selection (R S) problem with the goal to sel...

Ranking and Selection with Covariates for Personalized Decision Making

We consider a ranking and selection problem in the context of personaliz...

Please sign up or login with your details

Forgot password? Click here to reset