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:20260522T150124Z
LOCATION:D167/174
DTSTART;TZID=America/Chicago:20181111T143000
DTEND;TZID=America/Chicago:20181111T150000
UID:submissions.supercomputing.org_SC18_sess221_ws_mlhpce128@linklings.com
SUMMARY:Deep Learning Evolutionary Optimization for Regression of Rotorcra
 ft Vibrational Spectra
DESCRIPTION:Daniel A. Martinez-Gonzalez (US Army Engineer Research and Dev
 elopment Center) and Wesley Brewer (US Department of Defense HPC Moderniza
 tion Program)\n\nA method for Deep Neural Network (DNN) hyperparameter sea
 rch using evolutionary optimization is proposed for nonlinear high-dimensi
 onal multivariate regression problems. Deep networks often lead to extensi
 ve hyperparameter searches which can become an ambiguous process due to ne
 twork complexity. Therefore, we propose a user-friendly method that integr
 ates Dakota optimization library, TensorFlow, and Galaxy HPC workflow mana
 gement tool to deploy massively parallel function evaluations in a Genetic
  Algorithm (GA). Deep Learning Evolutionary Optimization (DLEO) is the cur
 rent GA implementation being presented. Compared with random generated and
  hand-tuned models, DLEO proved to be significantly faster and better sear
 ching for optimal architecture hyperparameter configurations. Implementing
  DLEO allowed us to find models with higher validation accuracies at lower
  computational costs in less than 72 hours, as compared with weeks of manu
 al and random search. Moreover, parallel coordinate plots provided valuabl
 e insights about network architecture designs and their regression capabil
 ities\n\nTag: Applications, Deep Learning, Machine Learning\n\nRegistratio
 n Category: Workshop Reg Pass\n\n
END:VEVENT
END:VCALENDAR