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:20210916T132527Z
LOCATION:Henry Dunant
DTSTART;TZID=Europe/Stockholm:20210706T140000
DTEND;TZID=Europe/Stockholm:20210706T160000
UID:submissions.pasc-conference.org_PASC21_sess136@linklings.com
SUMMARY:Artificial Intelligence Enabled Multiscale Molecular Simulations i
 n Biological and Material Sciences
DESCRIPTION:Minisymposium\n\nSimulations of physical phenomena consume a l
 arge fraction of computer time on existing supercomputing resources. Today
 , the challenge of scaling multiscale simulations is primarily addressed b
 y brute-force search-and-sample techniques and are computationally expensi
 ve. Emerging Exascale architectures pose challenges for simulations such a
 s efficient and scalable execution of complex workflows, concurrent execut
 ion of heterogeneous tasks and the robustness of algorithms on millions of
  processing cores, data and I/O parallelism, and fault tolerance. Therefor
 e, incremental approaches that scale simulations will not be successful in
  achieving the throughput and utilization on such machines. Machine learni
 ng (ML) techniques can be integrated with system and application changes a
 nd give many orders of magnitude higher effective performance. We term thi
 s convergence of high-performance computing (HPC) and ML methodologies/ pr
 actice as MLforHPC. Nowhere is the impact of MLforHPC methods likely to be
  greater than multiscale simulations in biological and material sciences, 
 with early evidence suggesting several orders of magnitude improvement ove
 r traditional methods. Fueled by advances in statistical algorithms and ru
 ntime systems, ensemble-based methods have overcome some of the limitation
 s of traditional monolithic simulations. Furthermore, integrating ML appro
 aches with such ensemble methods holds even greater promise in overcoming 
 performance barriers and enabling simulations of complex multiscale phenom
 ena.\n\nMarkov State Model Based Adaptive Sampling Algorithms\n\nShukla\n\
 nOne of the key limitations of Molecular Dynamics (MD) simulations is the 
 computational intractability of sampling protein conformational lands
 capes associated with either large system size or long time scales. To ove
 rcome this bottleneck, Markov State Model based adaptive sampling algorith
 ms ha...\n\n---------------------\nDistributed Deep Reinforcement Learning
  of Molecular Mechanisms\n\nCovino, Jung, Wadhawan, Bolhuis, Hummer\n\nWe 
 present a deep reinforcement learning artificial intelligence (AI) that le
 arns the mechanism of self-organization in molecular systems from computer
  simulations. The AI initiates molecular dynamics simulations of molecular
  reorganizations and progressively learns how to predict their outcome. We
  ...\n\n---------------------\nDesigning Molecular Models by Machine Learn
 ing and Experimental Data\n\nClementi\n\nThe last years have seen an immen
 se increase in high-throughput and high-resolution technologies for experi
 mental observation as well as high-performance techniques to simulate mole
 cular systems at a microscopic level, resulting in vast and ever-increasin
 g amounts of high-dimensional data. However, ...\n\n---------------------\
 nPanel Discussion: How AI and Multi Scale Simulations are Changing the Lan
 dscape of Bio-Molecular and Materials Systems\n\nRamanathan\n\nThe final s
 ession of our session will include a panel discussion amongst the presente
 rs and how they see the evolution of AI/ML strategies for the next generat
 ion of multiscale simulations. Audience will have an opprotunity to intera
 ct with the panel and discuss some emerging challenges in this area...\n\n
 \nDomain: Chemistry and Materials, Physics, Life Sciences, Engineering
END:VEVENT
END:VCALENDAR