BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.earthmonitor.org//global-workshop-2026//7MRGZM
BEGIN:VTIMEZONE
TZID:Europe/Amsterdam
BEGIN:STANDARD
DTSTART:20001029T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T020000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-global-workshop-2026-GCLJMU@pretalx.earthmonitor.org
DTSTART;TZID=Europe/Amsterdam:20261009T110000
DTEND;TZID=Europe/Amsterdam:20261009T113000
DESCRIPTION:Earth Observation is entering an era of abundance—global sate
 llite archives\, growing in-situ networks\, and an expanding open-source e
 cosystem—yet turning this distributed wealth into decision-ready environ
 mental information remains difficult because data are heterogeneous\, inco
 mplete\, and hard to combine across sensors\, resolutions\, and regions. T
 his keynote outlines Earth Embeddings: compact\, AI-native “mental maps
 ” that summarize what makes a location unique by learning directly from 
 imagery and context. Using an intuitive “Satellite GeoGuessr” contrast
 ive training setup\, neural networks learn place-specific visual and conte
 xtual signatures and distill them into dense vectors that can serve as por
 table location tokens in downstream models\, enabling reuse across tasks a
 nd regions. This talk will give an overview over different strategies to g
 enerate embeddings and outline research gaps and steps forward towards glo
 bal interoperable FAIR embeddings.
DTSTAMP:20260624T071620Z
LOCATION:Aula Magna
SUMMARY:Earth Embeddings: Learning “Mental Maps” for Open\, Interoperab
 le GeoAI - Marc Rußwurm
URL:https://pretalx.earthmonitor.org/global-workshop-2026/talk/GCLJMU/
END:VEVENT
END:VCALENDAR