skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Selecting the selector: Comparison of update rules for discrete global optimization

Abstract

In this paper, we compare some well-known Bayesian global optimization methods in four distinct regimes, corresponding to high and low levels of measurement noise and to high and low levels of “quenched noise” (which term we use to describe the roughness of the function we are trying to optimize). We isolate the two stages of this optimization in terms of a “regressor,” which fits a model to the data measured so far, and a “selector,” which identifies the next point to be measured. Finally, the focus of this paper is to investigate the choice of selector when the regressor is well matched to the data.

Authors:
ORCiD logo [1];  [2]
  1. Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
  2. Texas A & M Univ., Corpus Christi, TX (United States). Dept. of Mathematics and Statistics
Publication Date:
Research Org.:
Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
Sponsoring Org.:
USDOE; LANL Laboratory Directed Research and Development (LDRD) Program
OSTI Identifier:
1402614
Report Number(s):
LA-UR-16-24209
Journal ID: ISSN 1932-1864
Grant/Contract Number:
AC52-06NA25396
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
Statistical Analysis and Data Mining
Additional Journal Information:
Journal Volume: 10; Journal Issue: 4; Journal ID: ISSN 1932-1864
Publisher:
Wiley
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; optimization; Bayesian

Citation Formats

Theiler, James, and Zimmer, Beate G. Selecting the selector: Comparison of update rules for discrete global optimization. United States: N. p., 2017. Web. doi:10.1002/sam.11343.
Theiler, James, & Zimmer, Beate G. Selecting the selector: Comparison of update rules for discrete global optimization. United States. doi:10.1002/sam.11343.
Theiler, James, and Zimmer, Beate G. 2017. "Selecting the selector: Comparison of update rules for discrete global optimization". United States. doi:10.1002/sam.11343.
@article{osti_1402614,
title = {Selecting the selector: Comparison of update rules for discrete global optimization},
author = {Theiler, James and Zimmer, Beate G.},
abstractNote = {In this paper, we compare some well-known Bayesian global optimization methods in four distinct regimes, corresponding to high and low levels of measurement noise and to high and low levels of “quenched noise” (which term we use to describe the roughness of the function we are trying to optimize). We isolate the two stages of this optimization in terms of a “regressor,” which fits a model to the data measured so far, and a “selector,” which identifies the next point to be measured. Finally, the focus of this paper is to investigate the choice of selector when the regressor is well matched to the data.},
doi = {10.1002/sam.11343},
journal = {Statistical Analysis and Data Mining},
number = 4,
volume = 10,
place = {United States},
year = 2017,
month = 5
}

Journal Article:
Free Publicly Available Full Text
This content will become publicly available on May 24, 2018
Publisher's Version of Record

Save / Share:
  • In some areas of discrete optimization, it is necessary to isolate classes of problems whose target functions do not have local or strictly local minima that differ from the global minima. Examples include optimizations on discrete metric spaces and graphs, lattices and partially ordered sets, and linear combinatorial problems. A unified schema that to a certain extent generalizes the convexity models on which the above-cited works are based has been presented in articles. This article is a continuation of that research.
  • There is an increasing interest in the use of computer algorithms to identify combinations of parameters which optimise the energy performance of buildings. For such problems, the objective function can be multi-modal and needs to be approximated numerically using building energy simulation programs. As these programs contain iterative solution algorithms, they introduce discontinuities in the numerical approximation to the objective function. Metaheuristics often work well for such problems, but their convergence to a global optimum cannot be established formally. Moreover, different algorithms tend to be suited to particular classes of optimization problems. To shed light on this issue we comparedmore » the performance of two metaheuristics, the hybrid CMA-ES/HDE and the hybrid PSO/HJ, in minimizing standard benchmark functions and real-world building energy optimization problems of varying complexity. From this we find that the CMA-ES/HDE performs well on more complex objective functions, but that the PSO/HJ more consistently identifies the global minimum for simpler objective functions. Both identified similar values in the objective functions arising from energy simulations, but with different combinations of model parameters. This may suggest that the objective function is multi-modal. The algorithms also correctly identified some non-intuitive parameter combinations that were caused by a simplified control sequence of the building energy system that does not represent actual practice, further reinforcing their utility.« less
  • The radial Schroedinger equation with Coulomb potential perturbed on a compact set is considered. Estimates for the regular and singular solutions and a principle for selecting the eigenvalues are proved.