BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250626T234543Z
LOCATION:B306
DTSTART;TZID=America/New_York:20241117T111000
DTEND;TZID=America/New_York:20241117T113000
UID:submissions.supercomputing.org_SC24_sess734_ws_pawatm109@linklings.com
SUMMARY:Applying a Task-Based Approach to Distributed Machine Learning Wor
 kflows
DESCRIPTION:Fernando Vazquez-Novoa, Daniele Lezzi, Francesc Lordan, Fateme
 h Baghdadi, and Davide Cirillo (Barcelona Supercomputing Center (BSC))\n\n
 The growing demands across various scientific fields\nhave led to a signif
 icant shift in applications that consume data at\nthe edge of the computin
 g continuum. These applications require\nunified programming models for th
 e composition of components\nand coordinating the execution of computation
 al workloads,\nincluding training machine learning (ML) models on distribu
 ted\nresources. Personalized healthcare often leverages data generated\nfr
 om wearable devices used to train ML models, can be benefited\nfrom distri
 buted computing approaches. Specifically, stroke care\ncan be greatly bene
 fited from distributed ML with modifiable\nrisk factors that can be monito
 red using wearable devices. In this\nwork, we present an implementation th
 at leverages distributed\ntechniques for large-scale ML workflows using el
 ectrocardiogram\n(ECG) recordings for atrial fibrillation (AF) classificat
 ion. The\napplication was evaluated using the PhysioNet database, show-\nc
 asing the potential of distributed, ML in stroke care, opening\nthe way fo
 r future creation of more advanced models embedded\nin edge devices.\n\nTa
 g: Heterogeneous Computing, Parallel Programming Methods, Models, Language
 s and Environments, PAW-Full, Task Parallelism\n\nRegistration Category: W
 orkshop Reg Pass\n\nSession Chairs: Engin Kayraklioglu (Hewlett Packard En
 terprise (HPE)); Daniele Lezzi (Barcelona Supercomputing Center (BSC)); Ka
 rla Vanessa Morris Wright (Sandia National Laboratories); Irene Moulitsas 
 (Cranfield University); Elliott Slaughter (SLAC National Accelerator Labor
 atory); and Kenjiro Taura (The University of Tokyo, Japan)\n\n
END:VEVENT
END:VCALENDAR