BEGIN:VCALENDAR
VERSION:2.0
PRODID:www.dresden-science-calendar.de
METHOD:PUBLISH
CALSCALE:GREGORIAN
X-MICROSOFT-CALSCALE:GREGORIAN
X-WR-TIMEZONE:Europe/Berlin
BEGIN:VTIMEZONE
TZID:Europe/Berlin
X-LIC-LOCATION:Europe/Berlin
BEGIN:DAYLIGHT
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
DTSTART:19810329T030000
RRULE:FREQ=YEARLY;INTERVAL=1;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
DTSTART:19961027T030000
RRULE:FREQ=YEARLY;INTERVAL=1;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:DSC-22919
DTSTART;TZID=Europe/Berlin:20260610T130000
SEQUENCE:1781069903
TRANSP:OPAQUE
DTEND;TZID=Europe/Berlin:20260610T140000
URL:https://dresden-science-calendar.org/calendar/de/detail/22919
LOCATION:IFW\, Helmholtzstraße 2001069 Dresden
SUMMARY:Eberl: Materials data science and AI in Materials Science and Engin
 eering with a focus on micromechanics
CLASS:PUBLIC
DESCRIPTION:Speaker: Prof. Dr. Chris Eberl\nInstitute of Speaker: Fraunhofe
 r IWM\, Freiburg\nTopics:\n\n Location:\n  Name: IFW (D2E.27\, IFW Dresden
 )\n  Street: Helmholtzstraße 20\n  City: 01069 Dresden\n  Phone: \n  Fax:
  \nDescription: The advancement of available AI models allows to tackle di
 fferent challenges in Materials Science and Engineering. In this talk\, ap
 plications of different AI models shall be discussed in the context of mic
 romechanics. This allows to accelerate scientific workflows and automate t
 asks. The following examples will be discussed: (1) materials selection vi
 a simple models\, (2) classification and segmentation for microstructure a
 nalysis as well as damage identification and quantification via deep learn
 ing\, (3) fatigue lifetime prediction via graph neural networks and explai
 nable AI approaches based on structured data from high-throughput small sc
 ale testing\, (4) data and information extraction from publications via La
 rge Language Models and (5) semantic extraction and ontology learning for 
 MSE via agentic AI. While AI accelerates and enriches our understanding of
  materials\, reliability and uncertainties need to be critically reviewed.
  Therefore\, training data\, training methodologies as well as model selec
 tion are all relevant ingrediencies and will be discussed. The talk will c
 onclude by introducing national and international initiatives: NFDI\, NFDI
 -MatWerk (since 2021)\, MaterialDigital (since 2019)\, and Materials Commo
 ns (2026 - 2030 EU) and how to participate.
DTSTAMP:20260618T094432Z
CREATED:20260527T053710Z
LAST-MODIFIED:20260610T053823Z
END:VEVENT
END:VCALENDAR