BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Denver
X-LIC-LOCATION:America/Denver
BEGIN:DAYLIGHT
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
TZNAME:MDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
TZNAME:MST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260422T000711Z
LOCATION:506
DTSTART;TZID=America/Denver:20231113T165700
DTEND;TZID=America/Denver:20231113T170100
UID:submissions.supercomputing.org_SC23_sess454_ws_eduhpca101@linklings.co
 m
SUMMARY:Using MPI For Distributed Hyper-Parameter Optimization and Uncerta
 inty Evaluation
DESCRIPTION:Maria Pantoja (California Polytechnic State University, San Lu
 is Obispo; Argonne National Laboratory (ANL)); Erik Pautsch (Loyola Univer
 sity, Chicago); John Li (University of California, San Diego (UCSD)); Silv
 io Rizzi (Argonne National Laboratory (ANL)); and George Thiruvathukal (Lo
 yola University, Chicago)\n\nDeep Learning (DL) methods have recently domi
 nated the fields of Machine Learning. Most DL models assume that the input
  data distribution is identical between testing and validation, though the
 y often are not. For example, if we train a traffic sign classifier, the m
 odel might confidently, but incorrectly, classify a graffitied stop sign a
 s a speed limit sign. Often ML provides high-confidence (softmax) output f
 or out-of-distribution input that should have been classified as "I don't 
 know".  By adding the capability of propagating uncertainty to our results
 , the model can provide not just a single prediction, but a distribution o
 ver predictions that will allow the user to determine the model's reliabil
 ity and whether it needs to be deferred to a human expert. Uncertainty est
 imation is computationally expensive; in this assignment, we will learn to
  accelerate the calculations using common distributed systems divide and c
 onquer techniques.\n\nFiles given to students (Slides&Code ) (link:\url{ht
 tps://drive.google.com/drive/folders/1KrxWlMZpoJzph0Y7VbZj_yYyACK-Jusl?usp
 =sharing})\n\nTag: Education, Heterogeneous Computing, Reproducibility, St
 ate of the Practice, Sustainability\n\nRegistration Category: Workshop Reg
  Pass\n\nSession Chairs: David P. Bunde (Knox College); Sushil K. Prasad (
 University of Texas at San Antonio); Satish Puri (Missouri University of S
 cience and Technology); Apan Qasem (Texas State University); and George K.
  Thiruvathukal (Loyola University, Chicago; Argonne National Laboratory (A
 NL))\n\n
END:VEVENT
END:VCALENDAR