BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20181221T160726Z
LOCATION:D220
DTSTART;TZID=America/Chicago:20181111T142100
DTEND;TZID=America/Chicago:20181111T142400
UID:submissions.supercomputing.org_SC18_sess160_ws_whpc107@linklings.com
SUMMARY:Deep Learning: Extrapolation Tool for Computational Nuclear Physi
cs
DESCRIPTION:Workshop\nDiversity, Education, Hot Topics, Workshop Reg Pass\
n\nDeep Learning: Extrapolation Tool for Computational Nuclear Physics\n\
nNegoita\n\nThe goal of nuclear theory is to understand how nuclei arise f
rom interacting nucleons based on the underlying theory of the strong inte
ractions, quantum chromodynamics (QCD). The interactions among the nucleon
s inside a nucleus are dominated by the strong interaction, which is non-p
erturbative in the low-energy regime relevant for nuclear physics. With ac
cess to powerful High Performance Computing (HPC) systems, several ab init
io approaches have been developed to study nuclear structure and reactions
, such as the No-Core Shell Model (NCSM). The NCSM and other approaches r
equire an extrapolation of the results obtained in a finite basis space to
the infinite basis space limit and assessment of the uncertainty of those
extrapolations. Each observable requires a separate extrapolation and mo
st observables have no proven extrapolation method at the present time. We
propose a feed-forward artificial neural network (ANN) method as an extra
polation tool to obtain the ground state energy and the ground state point
proton root-mean-square (rms) radius and their extrapolation uncertaintie
s. We have generated theoretical data for 6Li by performing ab initio NCSM
calculations using basis spaces up through the largest computationally fe
asible basis space. The designed ANNs are sufficient to produce results fo
r these two very different observables in ^6Li from the ab initio NCSM res
ults in small basis spaces that satisfy the following theoretical physics
condition: independence of basis space parameters in the limit of extremel
y large matrices. Comparisons of the ANN results with other extrapolation
methods are also provided.
URL:https://sc18.supercomputing.org/presentation/?id=ws_whpc107&sess=sess1
60
END:VEVENT
END:VCALENDAR