BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
X-LIC-LOCATION:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20240626T180034Z
LOCATION:Level 2 Lobby
DTSTART;TZID=America/Los_Angeles:20240626T180000
DTEND;TZID=America/Los_Angeles:20240626T190000
UID:dac_DAC 2024_sess237_RESEARCH1592@linklings.com
SUMMARY:Adaptive Graph Learning for Efficient Thermal Analysis of the Chip
 let System under Interface Variations
DESCRIPTION:Work-in-Progress Poster\n\nJingbo Sun and Zhenyu Wang (Arizona
  State University), Frank Liu (Oak Ridge National Laboratory), Vidya A. Ch
 habria (Arizona State University), and Yu Cao (University of Minnesota)\n\
 nIn thermal analysis of a chiplet system, conventional numerical methods o
 r machine learning-based surrogate models face tremendous challenges in co
 mputation cost and accuracy, especially in the presence of process and mat
 erial variations. We propose Graph Neural Networks (GNNs) as a mathematica
 l framework for efficient and robust thermal analysis with composite mater
 ials. By modeling each region and their thermal interactions as a graph, w
 e continually adapt the GNN model under thermal interface variations. We v
 alidate our approach with numerical solutions and real thermal images from
  a crossbar unit, and demonstrate its speedup and accuracy in a 2.5D chipl
 et system.\n\nTopic: AI, Autonomous Systems, Cloud, Design, EDA, Embedded 
 Systems, IP, Security
END:VEVENT
END:VCALENDAR