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:20181221T160903Z
LOCATION:C2/3/4 Ballroom
DTSTART;TZID=America/Chicago:20181115T083000
DTEND;TZID=America/Chicago:20181115T170000
UID:submissions.supercomputing.org_SC18_sess324_post133@linklings.com
SUMMARY:Machine Learning for Adaptive Discretization in Massive Multiscale
  Biomedical Modeling
DESCRIPTION:Poster\nTech Program Reg Pass, Exhibits Reg Pass\n\nMachine Le
 arning for Adaptive Discretization in Massive Multiscale Biomedical Modeli
 ng\n\nHan, Gupta, Zhang, Bluestein, Deng\n\nFor multiscale problems, tradi
 tional time stepping algorithms use a single smallest time stepsize in ord
 er to capture the finest details; using this scale leads to a significant 
 waste of computing resources for simulating coarse-grained portion of the 
 problem. To improve computing efficiency for multiscale modeling, we propo
 se a novel state-driven adaptive time stepping (ATS) algorithm to automati
 cally adapt the time stepsizes to the underlying biophysical phenomena at 
 multiple scales. In this, we use a machine-learning based solution framewo
 rk to classify and label these states for regulating the time stepsizes. W
 e demonstrate the values of our ATS algorithm by assessing the accuracy an
 d efficiency of a multiscale two-platelet aggregation simulation. By compa
 ring with traditional algorithm for this simulation, our ATS algorithm sig
 nificantly improves the efficiency while maintaining accuracy. Our novel A
 TS algorithm presents a more efficient framework for solving massive multi
 scale biomedical problems.
URL:https://sc18.supercomputing.org/presentation/?id=post133&sess=sess324
END:VEVENT
END:VCALENDAR