BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Erlangen Centre for Astroparticle Physics - ECPv6.2.3.2//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Erlangen Centre for Astroparticle Physics
X-ORIGINAL-URL:https://ecap.nat.fau.de
X-WR-CALDESC:Events for Erlangen Centre for Astroparticle Physics
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:Europe/Berlin
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20200329T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20201025T010000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20200702T120000
DTEND;TZID=Europe/Berlin:20200702T130000
DTSTAMP:20260416T004634
CREATED:20200617T100717Z
LAST-MODIFIED:20200617T100717Z
UID:3046-1593691200-1593694800@ecap.nat.fau.de
SUMMARY:ECAP Seminar: Muhammad Kasim
DESCRIPTION:Up to two billion times acceleration of scientific simulations with deep neural architecture search\nComputer simulations are invaluable tools for scientific discovery. However\, accurate simulations are often slow to execute\, which limits their applicability to extensive parameter exploration\, large-scale data analysis\, and uncertainty quantification. A promising route to accelerate simulations by building fast emulators with machine learning requires large training datasets\, which can be prohibitively expensive to obtain with slow simulations. Here we present a method based on neural architecture search to build accurate emulators even with a limited number of training data. The method successfully accelerates simulations by up to 2 billion times in 10 scientific cases including astrophysics\, climate science\, biogeochemistry\, high energy density physics\, fusion energy\, and seismology\, using the same super-architecture\, algorithm\, and hyperparameters. Our approach also inherently provides emulator uncertainty estimation\, adding further confidence in their use. We anticipate this work will accelerate research involving expensive simulations\, allow more extensive parameters exploration\, and enable new\, previously unfeasible computational discovery.
URL:https://ecap.nat.fau.de/index.php/event/ecap-seminar-2020-07-02-muhammad-kasim/
LOCATION:Zoom
CATEGORIES:Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20200730T120000
DTEND;TZID=Europe/Berlin:20200730T130000
DTSTAMP:20260416T004634
CREATED:20200617T101240Z
LAST-MODIFIED:20200617T101240Z
UID:3049-1596110400-1596114000@ecap.nat.fau.de
SUMMARY:ECAP Seminar: Anita Reimer
DESCRIPTION:Identifying sources of high-energy neutrinos of the AGN type: A theoretical approach\nActive galactic nuclei (AGN) have long been predicted to emit neutrinos if they host sites of cosmic-ray acceleration to very high energies. Until a few years ago neutrino astrophysics was merely a prediction by (some) cosmic-ray theorists. It became reality with the first discovery of an astrophysical neutrino flux ~7 years ago. Although the origin of these neutrinos remains unclear up to now\, jets of AGN remain among the prime candidate sources.\nIn this presentation I will discuss the multimessenger approach in the framework of hadronic AGN emission models from a theoretical perspective. Special emphasis will be given on external target photon fields. \n  \n 
URL:https://ecap.nat.fau.de/index.php/event/ecap-seminar-2020-07-30-anita-reimer/
LOCATION:Zoom
CATEGORIES:Seminar
END:VEVENT
END:VCALENDAR