Solid Earth Sciences (S)
Session Sub-categoryComplex & General(CG)
Session IDS-CG62
TitlePotentiality of Machine Learning in Solid Earth Sciences
Short TitleMachine Learning in Solid Earth Sciences
Main Convener NameTakahiko Uchide
AffiliationResearch Institute of Earthquake and Volcano Geology, Geological Survey of Japan, National Institute of Advanced Industrial Science and Technology (AIST)
Co-Convener 1NameHirokuni Oda
AffiliationInstitute of Geology and Geoinformation, Geological Survey of Japan, AIST
Session LanguageJ
ScopeThe recent development of the machine learning techniques including the deep learning is leading innovations in various fields. The applications of these techniques to solid earth sciences are also expected to develop new frontiers by, for example, the classification, the pattern recognition, and the regression of data. On the other hand, there are concerns on the human-interpretability of the deep learning.
This session will provide an opportunity to share the application studies to various fields in solid earth science and inspire each other. We also look forward to studies addressing the black box issue of the deep learning.
Presentation FormatOral and Poster presentation
Invited AuthorsGregory C Beroza (Stanford University)
Tsuyoshi Murata (Tokyo Institute of Technology)
TimePresentation NoTitlePresenterAbstract
Oral Presentation May 26 AM1
09:00 - 09:20SCG62-01The Promise of Machine Learning in Solid Earth Geoscience: Examples from SeismologyGregory C Beroza
09:20 - 09:35SCG62-02Automatic Approach to Low-Frequency Earthquakes Detection in Southwest Japan Based on Deep Learning TechniqueYu Junyu
09:35 - 09:50SCG62-03Broadband ground motion waveform synthesis utilizing AI-based upsampling technique (2)Takahiro Maeda
09:50 - 10:05SCG62-04Forecast trial of postseismic deformetion after the 2011 Tohoku-Oki earthquake by recurrent neural networkNorifumi Yamaga
10:05 - 10:25SCG62-05Deep Learning Approaches for Eruption Prediction of SakurajimaTsuyoshi Murata
Presentation NoTitlePresenterAbstract
Poster Presentation May 26 PM2
SCG62-P01Earthquake Magnitude Prediction with Seismic Nucleation Phase based on Machine LearningKangming Wu
SCG62-P02Automatic signal classification for continuous seismic waveform records based on an unsupervised learning algorithm: Application to artificial noises and low-frequency tremorsYuki Kodera
SCG62-P03Detection of Gas Bubble signals recorded at the OBS Stations by Machine-Learningemmy TY CHANG
SCG62-P04Automatic Phase Picking of Seismograms by Deep Neural NetworkTakahiko Uchide
SCG62-P05Seismic event detection in time-frequency domain with CNN and phase picking with HEDLiping Fan
SCG62-P06Broadband ground-motion synthesis using embeddig machine learningTomohisa Okazaki
SCG62-P07LSTM and CNN Applications to Forecast Earthquake Magnitude Probability DistributionQi Liu
SCG62-P08FORCsensei: A machine learning framework to estimate optimized first-order reversal curve distributionsHirokuni Oda