Online Appendix for 'Efficient Learning for Selecting Top-m Context-Dependent Designs'

23 Pages Posted: 4 Jan 2023 Last revised: 19 Dec 2023

See all articles by Gongbo Zhang

Gongbo Zhang

Guanghua School of Management, Peking University

Sihua Chen

Independent

Kuihua Huang

affiliation not provided to SSRN

Yijie Peng

Peking University

Date Written: December 25, 2022

Abstract

Online Appendix for "Efficient Learning for Selecting Top-m Context-Dependent Designs", which is submitted to the IEEE Transactions on Automation Science and Engineering. We consider a simulation optimization problem for a context-dependent decision-making, which aims to determine the top-m designs for all contexts. Under a Bayesian framework, we formulate the optimal dynamic sampling decision as a stochastic dynamic programming problem, and develop a sequential sampling policy to efficiently learn the performance of each design under each context. The asymptotically optimal sampling ratios are derived to attain the optimal large deviations rate of the worst-case of probability of false selection. The proposed sampling policy is proved to be consistent and its asymptotic sampling ratios are asymptotically optimal. Numerical experiments demonstrate that the proposed method improves the efficiency for selection of top-m context-dependent designs.

Keywords: Simulation optimization; context-dependent decision; top-m selection; dynamic sampling; asymptotic optimality.

Suggested Citation

Zhang, Gongbo and Chen, Sihua and Huang, Kuihua and Peng, Yijie, Online Appendix for 'Efficient Learning for Selecting Top-m Context-Dependent Designs' (December 25, 2022). Available at SSRN: https://ssrn.com/abstract=4316394 or http://dx.doi.org/10.2139/ssrn.4316394

Gongbo Zhang (Contact Author)

Guanghua School of Management, Peking University ( email )

Sihua Chen

Independent

Kuihua Huang

affiliation not provided to SSRN

Yijie Peng

Peking University ( email )

No 5 Yiheyuan Rd
Haidian District
Beijing, Beijing 100871
China

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
24
Abstract Views
137
PlumX Metrics