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:20211208T164412Z
LOCATION:3018
DTSTART;TZID=America/Los_Angeles:20211208T163000
DTEND;TZID=America/Los_Angeles:20211208T165000
UID:dac_DAC 2021_sess134_RESEARCH1128@linklings.com
SUMMARY:EMGraph: Fast Learning-Based Electromigration Analysis for Multi-S
 egment Interconnect Using Graph Convolution Networks
DESCRIPTION:Research Manuscript\n\nEMGraph: Fast Learning-Based Electromig
 ration Analysis for Multi-Segment Interconnect Using Graph Convolution Net
 works\n\nJin\n\nElectromigration (EM) becomes a major concern for VLSI cir
 cuits as the technology advances in  nanometer regime. VLSI interconnect c
 an be naturally viewed as graphs. Based on this observation, we propose a 
 new graph convolution network (GCN) model, which is called EMGraph, to est
 imate transient EM stress. Compared with generative adversarial network (G
 AN) method, EMGraph model can learn more transferable knowledge to predict
  stress on new graphs without retraining via inductive learning. Experimen
 tal results show the model has 1.5% averaged error compared to the ground 
 truth. It also achieves smaller model size, 4× accuracy and 14× speedup ov
 er the GAN-based method.\n\nTopic: EDA\n\nKeyword: Manufacturing Test and 
 Reliability\n\nVirtual Program: Presented In-Person
END:VEVENT
END:VCALENDAR