Skoči na glavni sadržaj

Izvorni znanstveni članak

https://doi.org/10.17535/crorr.2018.0001

An analysis of covariance parameters in Gaussian process-based optimization

Hossein Mohammadi ; College of Engineering, Mathematics and Physical Sciences, University of Exeter, Exeter, UK
Rodolphe Le Riche ; Ecole des Mines de Saint Etienne, Institut H. Fayol, Saint-Etienne, France
Xavier Bay ; Ecole des Mines de Saint Etienne, Institut H. Fayol, Saint-Etienne, France
Eric Touboul ; Ecole des Mines de Saint Etienne, Institut H. Fayol, Saint-Etienne, France


Puni tekst: engleski pdf 1.091 Kb

str. 1-10

preuzimanja: 696

citiraj


Sažetak

The need for globally optimizing expensive-to-evaluate functions frequently occurs in many real-world applications. Among the methods developed for solving such problems, the Efficient Global Optimization (EGO) is regarded as one of the state-of-the-art unconstrained continuous optimization algorithms. The surrogate model used in EGO is a Gaussian process (GP) conditional on data points. The most important control on the efficiency of the EGO algorithm is the GP covariance function (or kernel), which is taken as a parameterized function. In this paper, we theoretically and empirically analyze the effect of the covariance parameters, the so-called "characteristic length scale" and "nugget", on EGO performance. More precisely, we analyze the EGO algorithm with fixed covariance parameters and compare them to the standard setting where they are statistically estimated. The limit behavior of EGO with very small or very large characteristic length scales is identified. Experiments show that a "small" nugget should be preferred to its maximum likelihood estimate. Overall, this study contributes to a better theoretical and practical understanding of a key optimization algorithm.

Ključne riječi

Covariance kernel; EGO; Gaussian process; global optimization

Hrčak ID:

203889

URI

https://hrcak.srce.hr/203889

Datum izdavanja:

24.7.2018.

Posjeta: 1.864 *