BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20241120T082409Z
LOCATION:HG E 3
DTSTART;TZID=Europe/Stockholm:20240604T110000
DTEND;TZID=Europe/Stockholm:20240604T130000
UID:submissions.pasc-conference.org_PASC24_sess144@linklings.com
SUMMARY:MS3E - In Situ Coupling of Simulations and AI/ML for HPC: Software
 , Methodologies, and Applications - Part I
DESCRIPTION:Minisymposium\n\nMotivated by the remarkable success of artifi
 cial intelligence (AI) and machine learning (ML) in the fields of computer
  vision and natural language processing, over the last decade there has be
 en a host of successful applications of AI/ML to a variety of scientific d
 omains. In most cases, the models are trained using the traditional offlin
 e (or post hoc) approach, wherein the training data is produced, assembled
 , and curated separately before training is deployed. While more straightf
 orward, the offline training workflow can impose some important restrictio
 ns to the adoption of ML models for scientific applications. To solve thes
 e limitations, in situ (or online) ML approaches, wherein ML tasks are per
 formed concurrently to the ongoing simulation, have recently emerged as an
  attractive new paradigm. In this minisymposium, we explore novel approach
 es to enable the coupling of state-of-the-art simulation codes with differ
 ent AI/ML techniques. We discuss the open-source software libraries that a
 re being developed to solve the software engineering challenges of in situ
  ML workflows, as well as the methodologies adopted to scale on modern HPC
  systems and their applications to solve complex problems in different com
 putational science domains.\n\nChallenges and Opportunities in Combining L
 LMs with Conventional Simulation Workflows\n\nThe effectiveness of AI for 
 a variety of scientific tasks has rapidly improved over the last few years
 , changing the way that we can perform scientific workflows on HPC. Our re
 cent work deploying a workflow around a Large-Language Model (LLM) for gen
 erating protein sequences is a good example of man...\n\n\nLogan Ward, Gau
 tham Dharuman, Arvind Ramanathan, and Ian Foster (Argonne National Laborat
 ory)\n---------------------\nRelexi: Reinforcement Learning for Applicatio
 ns in Computational Fluid Dynamics\n\nRelexi is a powerful tool that allow
 s to use existing simulation codes as training environments for reinforcem
 ent learning (RL) on high-performance computing (HPC) systems. This framew
 ork allows to apply RL to problems typically requiring HPC hardware such a
 s computational fluid dynamics (CFD) or re...\n\n\nMarius Kurz (University
  of Stuttgart), Philipp Offenhäuser (HPE), Benjamin Sanderse (Centrum Wisk
 unde & Informatica (CWI)), and Andrea Beck (University of Stuttgart)\n----
 -----------------\nA Technical Overview of SmartSim and its Use Cases\n\nS
 ince its release in 2021, SmartSim has been gaining momentum in several sc
 ientific domains, including climate modeling and CFD. SmartSim’s unique se
 t of features allow researchers to perform in-situ data analysis, extend c
 apabilities of well-established numerical software with cutting-edge AI...
 \n\n\nAlessandro Rigazzi and Andrew Shao (HPE)\n---------------------\nMac
 hine-Learning Emulation of the Radiative Transfer Module in a Surface Cont
 inental Model ORCHIDEE\n\nThe ORCHIDEE land surface model is one of the IP
 SL's Earth System Model components. The radiative transfer portion of the 
 land model calculates reflected, absorbed and transmitted light at multipl
 e canopy levels. This calculation is crucial to the climate system, but is
  also the most time-consuming ...\n\n\nXiaoni Wang (CNRS, Le Laboratoire d
 es Sciences du Climat et de l'Environnement); Andrew Shao (HPE); Mandresy 
 Rasolonjatovo (UVSQ, Le Laboratoire des Sciences du Climat et de l'Environ
 nement); Fabienne Maignan (CEA, Le Laboratoire des Sciences du Climat et d
 e l'Environnement); and Philippe Peylin (CNRS, Le Laboratoire des Sciences
  du Climat et de l'Environnement)\n\nDomain: Chemistry and Materials, Clim
 ate, Weather, and Earth Sciences, Engineering, Computational Methods and A
 pplied Mathematics\n\nSession Chair: Riccardo Balin (Argonne National Labo
 ratory)
END:VEVENT
END:VCALENDAR