Research about pruning hyper-parameter optimization method based on transfer learning in geographic information system

Recently, researchers have found that some automated optimization methods and techniques can speed up the whole search process and can obtain better model hyper-parameter configurations. In this paper, the optimized initial hyper-parameter configuration is obtained through the loss function defined...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Arabian journal of geosciences Jg. 14; H. 5
Hauptverfasser: Zhang, Xiaohang, Li, Yuqi, Li, Zhengren
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Cham Springer International Publishing 01.03.2021
Springer Nature B.V
Schlagworte:
Algorithms
Configurations
Design optimization
Design parameters
Earth and Environmental Science
Earth science
Earth Sciences
Geographic information systems
Geographical information systems
Geological Modeling and Geospatial Data Analysis
Gmgda 2019
Information systems
Machine learning
Mathematical models
Optimization
Parameters
Remote sensing
Search process
Transfer learning
Geographic information system (GIS)
Hyper-parameter optimization
Automated machine learning algorithm
Pruning strategy
Transfer learning
SMBO model
ISSN:1866-7511, 1866-7538
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Recently, researchers have found that some automated optimization methods and techniques can speed up the whole search process and can obtain better model hyper-parameter configurations. In this paper, the optimized initial hyper-parameter configuration is obtained through the loss function defined under the definition of meta-features combined with the algorithm of transfer learning. Furthermore, the sequence model-based hyper-parameter optimization model (SMBO model) is transformed by introducing a pruning strategy. By categorizing the hyper-parameters according to their importance, the non-important hyper-parameters are eliminated but important hyper-parameters are strengthened, so the final hyper-parameter optimization configuration is obtained. It is the parameter optimization design based on this secondary optimization that makes it possible to quickly search and select relevant information and carry out data application. The main goal of this paper is to clarify the mathematical principle of secondary optimization and combine it with geographic information system to create the value development of geographic information database.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1866-7511
1866-7538
DOI:10.1007/s12517-021-06465-0