BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//University of Iceland//AI Centre//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
UID:b21b6760-b498-4abe-ab97-3009ed0f60b3@ai.hi.is
DTSTAMP:20261011T043453Z
DTSTART:20251118T090000Z
DTEND:20251118T100000Z
SUMMARY:Emergent Statistical Laws at Scale: b-values\, Mainshocks\, and Syn
 thetic Data for ML
DESCRIPTION:Earthquake catalogs look messy\, yet they obey a striking regul
 arity: the Gutenberg–Richter (GR) law\, where the number of events drops
  roughly as 10^−bM with magnitude M. The slope b summarizes how often sm
 all versus large quakes occur and has been proposed as a real-time clue fo
 r distinguishing foreshocks from mainshocks\, making it critical for hazar
 d assessment during unfolding seismic sequencies. In this talk\, I’ll sh
 ow how very large fault-system simulations—driven by frictional physics 
 and elasticity—spontaneously reproduce GR-like behavior and exhibit info
 rmative shifts in the b value. Because these simulators are fully controll
 ed\, we can test such ideas causally rather than anecdotally. I’ll then 
 outline a path for ML: using these models to generate rich\, labeled synth
 etic datasets for prediction and feature discovery. A key “predictabilit
 y knob” is the ratio Linf​/L (nucleation length to system size)\, whic
 h lets us tune dynamics from highly irregular to quasi-periodic—providin
 g rigorous benchmarks for machine learning methods.\n\nhttps://ai.hi.is/is
 /events/colloquium-in-statistics-and-ai-2/
LOCATION:Íslensk erfðagreining\, Fróði
URL:https://ai.hi.is/is/events/colloquium-in-statistics-and-ai-2/
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR