BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//localz//Events//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
UID:event-179574@localz.events
DTSTAMP:20261122T085348Z
LAST-MODIFIED:20261120T053017Z
DTSTART:20261130T170000Z
DTEND:20261130T180000Z
SUMMARY:Geophysics Master's Presentation\, Hamad Alswaidan: "Machine Learni
 ng-Assisted Full Waveform Inversion: Low-Frequency Extrapolation and Cycle
 -Skipping Mitigation"
DESCRIPTION:Full waveform inversion (FWI) is a powerful technique for estim
 ating subsurface physical properties from seismic data. However\, its nonl
 inear nature and the limited availability of low-frequency seismic data ma
 ke it susceptible to cycle skipping\, often preventing convergence toward 
 an accurate subsurface model. In this work\, Hamad explores two complement
 ary machine learning approaches to address these challenges. The first inv
 estigates the use of generative diffusion models to extrapolate missing lo
 w-frequency content in seismic data. The second introduces a neural networ
 k-based approach to mitigate cycle skipping through the alignment of obser
 ved and modeled seismic waveforms during inversion. Together\, these appro
 aches explore how machine learning can complement physics-based inversion 
 and improve the robustness of FWI.\n\nhttps://localz.events/e/geophysics-m
 asters-presentation-hamad-alswaidan-machine-learning-assisted-full-wavefor
 m-inversion-low-frequency-extrapolation-and-cycle-skipping-mitigation-1795
 74
LOCATION:Mitchell Earth Sciences\, 397 Panama Mall\, Stanford\, CA 94305\, 
 Stanford
GEO:37.426402;-122.172635
URL:https://localz.events/e/geophysics-masters-presentation-hamad-alswaidan
 -machine-learning-assisted-full-waveform-inversion-low-frequency-extrapola
 tion-and-cycle-skipping-mitigation-179574
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR
