Evolving Gaussian process kernels from elementary mathematical expressions for time series extrapolation

NEUROCOMPUTING

卷:462

Article

Roman, Ibai^1,2  Santana, Roberto¹  Mendiburu, Alexander¹  Lozano, Jose A.^1,3 

[1] Univ Basque Country, UPV EHU, Intelligent Syst Grp, Paseo Manuel Lardizabal 1, Donostia San Sebastian 20018, Spain

[2] Mondragon Univ, Software & Syst Engn Grp, Loramendi,4, Arrasate Mondragon 20500, Spain

[3] BCAM, Basque Ctr Appl Math, Alameda Mazarredo 14, Bilbao 48009, Spain

关键词: Evolutionary search;    Gaussian processes;    Genetic programming;    Kernel learning;    Time series extrapolation;   

DOI  :  10.1016/j.neucom.2021.08.020

来源: Elsevier

【 摘 要 】

Choosing the best kernel is crucial in many Machine Learning applications. Gaussian Processes are a state-of-the-art technique for regression and classification that heavily relies on a kernel function. However, in the Gaussian Processes literature, kernels have usually been either ad hoc designed, selected from a predefined set, or searched for in a space of compositions of kernels which have been defined a priori. In this paper, we propose a Genetic Programming algorithm that represents a kernel function as a tree of elementary mathematical expressions. By means of this representation, a wider set of kernels can be modeled, where potentially better solutions can be found, although new challenges also arise. The proposed algorithm is able to overcome these difficulties and find kernels that accurately model the characteristics of the data. This method has been tested in several real-world time series extrapolation problems, improving the state-of-the-art results while reducing the complexity of the kernels. (c) 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http:// creativecommons.org/licenses/by/4.0/).

【 授权许可】

Free