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:20210402T160046Z
LOCATION:Track 2
DTSTART;TZID=America/New_York:20201117T110000
DTEND;TZID=America/New_York:20201117T113000
UID:submissions.supercomputing.org_SC20_sess179_pap233@linklings.com
SUMMARY:MeshfreeFlowNet: A Physics-Constrained Deep Continuous Space-Time 
 Super-Resolution Framework
DESCRIPTION:Paper\n\nMeshfreeFlowNet: A Physics-Constrained Deep Continuou
 s Space-Time Super-Resolution Framework\n\nJiang, Esmaeilzadeh, Azizzadene
 sheli, Kashinath, Mustafa...\n\nWe propose MeshfreeFlowNet, a novel deep l
 earning framework, to generate continuous (grid-free) spatiotemporal solut
 ions from the low-resolution inputs. While being computationally efficient
 , MeshfreeFlowNet accurately recovers the fine-scale quantities of interes
 t. MeshfreeFlowNet allows for: (i) the output to be sampled at all spatio-
 temporal resolutions; (ii) a  set of Partial Differential Equation (PDE) c
 onstraints to be imposed; and (iii) training on fixed-size inputs on arbit
 rarily sized spatiotemporal domains owing to its fully convolutional encod
 er.\n\nWe empirically study the performance of PCSR on the task of super-r
 esolution of turbulent flows in the Rayleigh–Bénard convection problem. Ac
 ross a diverse set of evaluation metrics, we show that PCSR significantly 
 outperforms existing baselines. Furthermore, we provide a large scale impl
 ementation of PCSR and show that it efficiently scales across large cluste
 rs, achieving 96.80% scaling efficiency on up to 128 GPUs and a training t
 ime of less than 4 minutes.\n\nTag: Applications, Machine Learning, Deep L
 earning and Artificial Intelligence, Parallel Programming Languages, Libra
 ries, and Models\n\nRegistration Category: Tech Program Reg Pass\n\nAward 
 Finalist: Best Student Paper Finalists
END:VEVENT
END:VCALENDAR