Recognition of geochemical anomalies using a deep autoencoder network

In this paper, we train an autoencoder network to encode and reconstruct a geochemical sample population with unknown complex multivariate probability distributions. During the training, small probability samples contribute little to the autoencoder network. These samples can be recognized by the tr...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Computers & geosciences Jg. 86; S. 75 - 82
Hauptverfasser: Xiong, Yihui, Zuo, Renguang
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Ltd 01.01.2016
Schlagworte:
Anomalies
Autoencoder network
case studies
China
computers
Deposition
Errors
Geochemical exploration
Geochemistry
iron
learning
Mineralization
Multivariate geochemical data
Networks
probability
probability distribution
Reconstruction
Skarn-type iron deposits
Training
China
Skarn-type iron deposits
Multivariate geochemical data
Autoencoder network
Geochemical exploration
ISSN:0098-3004, 1873-7803
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In this paper, we train an autoencoder network to encode and reconstruct a geochemical sample population with unknown complex multivariate probability distributions. During the training, small probability samples contribute little to the autoencoder network. These samples can be recognized by the trained model as anomalous samples due to their comparatively higher reconstructed errors. The southwestern Fujian district (China) is chosen as a case study area. A variety of learning rates, iterations, and the size of each hidden layer are constructing and training the deep autoencoder networks on all the geochemical samples. The reconstruction error (or, anomaly score) of each training sample is used to recognize multivariate geochemical anomalies associated with Fe polymetallic mineralization. By comparing the results obtained with a continuous restricted Boltzmann machine, we conclude that the autoencoder network can be trained to recognize multivariate geochemical anomalies. Most of the known skarn-type Fe deposits are located in areas with high reconstruction errors or anomaly scores in the anomaly map, indicating that these anomalies may be related to Fe mineralization. •Recognition of geochemical anomalies using an autoencoder network.•Recognition of geochemical anomalies using continuous restricted Boltzmann machines.•Methods demonstrated in a case study southwestern Fujian district (China).
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0098-3004
1873-7803
DOI:10.1016/j.cageo.2015.10.006