论文信息 - A Parallel Program for 3-Arm Bandits

A Parallel Program for 3-Arm Bandits

We describe a new parallel program for optimizing and analyzing 3-arm Bernoulli bandit problems. Previous researchers had considered this problem computationally intractable, and we know of no previous exact optimizations of 3-arm bandit problems. Despite this, our program is able to solve problems of size 100 or more. We describe the techniques used to achieve this, and indicate various extensions of the program that enable it to solve a wide range of related problems.

Quentin F. Stout | Janis Hardwick | Robert H. Oehmke

[1] R. Simon,et al. Adaptive treatment assignment methods and clinical trials. , 1977, Biometrics.

[2] Rebecca A. Betensky,et al. An O'Brien-Fleming sequential trial for comparing three treatments , 1996 .

[3] P. Armitage. The search for optimality in clinical trials , 1985 .

[4] V. G. Kulkarni,et al. Optimal Bayes procedures for selecting the better of two Bernoulli populations , 1986 .

[5] P. W. Jones,et al. Bandit Problems, Sequential Allocation of Experiments , 1987 .

[6] D. Coad. SEQUENTIAL ALLOCATION INVOLVING SEVERAL TREATMENTS , 1995 .

[7] Quentin F. Stout,et al. Optimal allocation for estimating the mean of a bivariate polynomial , 1996 .

[8] Quentin F. Stout,et al. Exact computational analyses for adaptive designs , 1995 .

[9] D. Berry,et al. Adaptive assignment versus balanced randomization in clinical trials: a decision analysis. , 1995, Statistics in medicine.

[10] You-Gan Wang. Sequential allocation in clinical trials , 1991 .