Geostatistical Learning is a new branch of Geostatistics concerned with learning functions over geospatial domains (e.g. 2D maps, 3D subsurface models). The theory is being carefully implemented in the GeoStats.jl framework, which is an extensible framework for high-performance geostatistics in Julia. In this talk, I will illustrate how the framework can be used to learn functions over general unstructured meshes, and how this unique technology can help advance geoscientific work.
The theory was introduced in our recent (open access) paper available online: https://www.frontiersin.org/articles/10.3389/fams.2021.689393/full
Its implementation requires knowledge of geostatistics, computational geometry, and high-performance computing. Due to the great features of the Julia language, we were able to achieve an elegant design with great runtime performance.
For more info on the Julia Programming Language, follow us on Twitter: https://twitter.com/JuliaLanguage and consider sponsoring us on GitHub: https://github.com/sponsors/JuliaLang
Contents
00:00 Welcome!
00:29 The two connotations of the word "Geo"
00:47 Here we understand GEOstatistics as statistics developed for GEOspatial data
01:02 Geospatial data is a combination of tables of attributes and discretization of the geospatial domain
01:12 We support any table implementing Table.jl interface
01:23 We support any domain implementing Meshes.jl interface
01:40 Makie.jl allows use to visualize these domains efficiently on GPU
02:06 Example 1: 3D grid data
03:38 Example 2: 2D grid data (a.k.a. image)
04:28 Example 3: Map data
05:16 Example 4: Mesh data
05:59 Classical learning framework
06:50 Assumptions of classical learning framework do NOT hold in GEOspatial applications
07:20 Problem 1: Why the error is so high?
09:02 Samples are geospatial correlated
09:27 Cross-validation (CV) vs geostatistical validation
11:15 Showcase of working code
12:22 Problem 2: Why the clusters are everywhere?
13:07 Geostatistical clustering methods
13:47 We propose a new framework: geostatistical learning
14:16 Advanced example: learning Wind-Chill Index (WCI) for models of airplanes and helicopters
15:30 Advanced example: Wind-Chill Index for a model of a helicopter
16:31 Advanced example: Final result
17:19 Challenges and opportunities
17:56 We invite you to join our community if you share our feeling about geostatistics and industry
S/O to https://github.com/KZiemian for the video timestamps!
Want to help add timestamps to our YouTube videos to help with discoverability? Find out more here: https://github.com/JuliaCommunity/YouTubeVideoTimestamps
Interested in improving the auto generated captions? Get involved here: https://github.com/JuliaCommunity/YouTubeVideoSubtitles
3 Comments