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:20240626T180035Z
LOCATION:3010\, 3rd Floor
DTSTART;TZID=America/Los_Angeles:20240626T144500
DTEND;TZID=America/Los_Angeles:20240626T150000
UID:dac_DAC 2024_sess138_RESEARCH243@linklings.com
SUMMARY:Nona: Accurate Power Prediction Model Using Neural Networks
DESCRIPTION:Research Manuscript\n\nHoSun Choi, Chanho Park, Euijun Kim, an
 d William Song (Yonsei University)\n\nThis paper proposes a neural-network
 -based power model, Nona, that accurately predicts the power consumption o
 f heterogeneous CPUs on a commercial mobile device. With aggressive on-dev
 ice power management in action, it becomes increasingly challenging to mak
 e accurate power predictions for diverse applications. To overcome the lim
 itations of the existing power models based on linear regression, Nona use
 s a lightweight neural network with a small number of performance monitori
 ng counters (PMCs) chosen from a system analysis and a loss function desig
 ned for power prediction.\nExperiments on Google Pixel 6 show that Nona ha
 s a 3.4% average prediction error, improving on prior work by 2.6x.\n\nTop
 ic: EDA\n\nKeyword: Timing and Power Analysis and Optimization\n\nSession 
 Chairs: Zhou Jin (China University of Petroleum) and Umamaheswara Rao Tida
  (North Dakota State University)
END:VEVENT
END:VCALENDAR