BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210402T160556Z
LOCATION:Track 7
DTSTART;TZID=America/New_York:20201111T100000
DTEND;TZID=America/New_York:20201111T183000
UID:submissions.supercomputing.org_SC20_sess199@linklings.com
SUMMARY:The 5th Deep Learning on Supercomputers Workshop
DESCRIPTION:Workshop\n\nDLS – Introduction: The 5th Deep Learning on Super
 computers Workshop\n\nZhang, Foster, Codreanu\n\nThe Deep Learning (DL) on
  Supercomputers workshop provides a forum for practitioners working on any
  and all aspects of DL for science and engineering in the high-performance
  computing (HPC) context to present their latest research results and deve
 lopment, deployment and application experiences. The ...\n\n--------------
 -------\nDLS – Break\n\n\n\n---------------------\nDLS – Break\n\n\n\n----
 -----------------\nDLS – Break\n\n\n\n---------------------\nDLS – Lunch B
 reak\n\n\n\n---------------------\nAI for Science: AI + HPC\n\nStevens\n\n
 ---------------------\nTowards a Scalable and Distributed Infrastructure f
 or Deep Learning Applications\n\nHasheminezhad, Shirzad, Wu, Diehl, Schulz
 ...\n\nAlthough recent scaling up approaches to train deep neural networks
  have proven to be effective, the computational intensity of large and com
 plex models, as well as the availability of large-scale datasets require d
 eep learning frameworks to utilize scaling out techniques. Parallelization
  approaches...\n\n---------------------\nDDLBench: Towards a Scalable Benc
 hmarking Infrastructure for Distributed Deep Learning\n\nJansen, Codreanu,
  Varbanescu\n\nDue to its many applications across various fields of resea
 rch, engineering and daily life, deep learning has seen a surge in popular
 ity. Therefore, larger and more expressive models have been proposed, with
  examples like Turing-NLG using as many as 17 billion parameters. Training
  these very large m...\n\n---------------------\nVandermonde Wave Function
  Ansatz for Improved Variational Monte Carlo\n\nAcevedo, Curry, Leroux, Jo
 shi, Malaya\n\nSolutions to the Schrödinger equation can be used to predic
 t the electronic structure of molecules and materials and therefore infer 
 their complex physical and chemical properties. Variational Quantum Monte 
 Carlo (VMC) is a technique that can be used to solve the weak form of the 
 Schrödinger equatio...\n\n---------------------\nOnline-Codistillation Mee
 ts LARS: Going beyond the Limit of Data Parallelism in Deep Learning\n\nMu
 rai, Mikami, Koyama, Suzuki, Akiba\n\nData parallel training is a powerful
  family of methods for the efficient training of deep neural networks on b
 ig data. Unfortunately, however, recent studies have shown that the merit 
 of increased batch-size in terms of both speed and model-performance dimin
 ishes rapidly beyond some point.  This see...\n\n---------------------\nTi
 me-Based Roofline for Deep Learning Performance Analysis\n\nWang, Yang, Fa
 rrell, Zhang, Kurth...\n\nDeep learning applications based on neural netwo
 rks are generating considerable interest in various fields due to their hi
 gh accuracy. Such an application is usually very compute-intensive and thu
 s requires a long run time. Researchers and engineers are actively explori
 ng new solutions to this issue...\n\n---------------------\nTopiQAL: Topic
 -aware Question Answering Using Scalable Domain-Specific Supercomputers\n\
 nVenkataram, Mattmann, Penberthy\n\nWe all have questions, about today's t
 emperature, scores of our favorite baseball team, the Universe, and about 
 life during COVID-19. Life, physical and natural scientists have been tryi
 ng to find answers to various topics using scientific methods and experime
 nts, while computer scientists have buil...\n\n---------------------\nDeep
 Galaxy: Deducing the Properties of Galaxy Mergers from Images Using Deep N
 eural Networks\n\nCai, Bédorf, Saletore, Codreanu, Podareanu...\n\nGalaxy 
 mergers, the dynamical process during which two galaxies collide, are amon
 g the most spectacular phenomena in the Universe. During this process, the
  two colliding galaxies are tidally disrupted, producing significant visua
 l features that evolve as a function of time. These visual features con...
 \n\n---------------------\nExploring the Limits of Concurrency in ML Train
 ing on Google TPUs\n\nKumar\n\n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR