eduzhai > Applied Sciences > Engineering >

70 years of machine learning in geoscience in review

  • Save

... pages left unread,continue reading

Document pages: 36 pages

Abstract: This review gives an overview of the development of machine learning ingeoscience. A thorough analysis of the co-developments of machine learningapplications throughout the last 70 years relates the recent enthusiasm formachine learning to developments in geoscience. I explore the shift of krigingtowards a mainstream machine learning method and the historic application ofneural networks in geoscience, following the general trend of machine learningenthusiasm through the decades. Furthermore, this chapter explores the shiftfrom mathematical fundamentals and knowledge in software development towardsskills in model validation, applied statistics, and integrated subject matterexpertise. The review is interspersed with code examples to complement thetheoretical foundations and illustrate model validation and machine learningexplainability for science. The scope of this review includes various shallowmachine learning methods, e.g. Decision Trees, Random Forests, Support-VectorMachines, and Gaussian Processes, as well as, deep neural networks, includingfeed-forward neural networks, convolutional neural networks, recurrent neuralnetworks and generative adversarial networks. Regarding geoscience, the reviewhas a bias towards geophysics but aims to strike a balance with geochemistry,geostatistics, and geology, however excludes remote sensing, as this wouldexceed the scope. In general, I aim to provide context for the recententhusiasm surrounding deep learning with respect to research, hardware, andsoftware developments that enable successful application of shallow and deepmachine learning in all disciplines of Earth science.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×