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:501-502
DTSTART;TZID=America/Denver:20231113T153000
DTEND;TZID=America/Denver:20231113T155000
UID:submissions.supercomputing.org_SC23_sess440_ws_ai4s123@linklings.com
SUMMARY:Tournament-Based Pretraining to Accelerate Federated Learning
DESCRIPTION:Matt Baughman (University of Chicago); Nathaniel Hudson (Unive
 rsity of Chicago, Argonne National Laboratory (ANL)); Ryan Chard (Argonne 
 National Laboratory (ANL)); Andre Bauer (University of Chicago); Ian Foste
 r (Argonne National Laboratory (ANL)); and Kyle Chard (University of Chica
 go, Argonne National Laboratory (ANL))\n\nAdvances in hardware, proliferat
 ion of compute at the edge, and data creation at unprecedented scales have
  made federated learning (FL) necessary for the next leap forward in perva
 sive machine learning. For privacy and network reasons, large volumes of d
 ata remain stranded on endpoints located in geographically austere (or at 
 least austere network-wise) locations. However, challenges exist to the ef
 fective use of these data. To solve the system and functional level challe
 nges, we present an three novel variants of a serverless federated learnin
 g framework. We also present tournament-based pre-training, which we demon
 strate significantly improves model performance in some experiments. Overa
 ll, these extensions to FL and our novel training method enable greater fo
 cus on science rather than ML development.\n\nTag: Artificial Intelligence
 /Machine Learning\n\nRegistration Category: Workshop Reg Pass\n\nSession C
 hairs: Murali Emani (Argonne National Laboratory (ANL)); Gokcen Kestor (Ba
 rcelona Supercomputing Center (BSC); University of California, Merced); an
 d Dong Li (University of California, Merced)\n\n
END:VEVENT
END:VCALENDAR