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:20260402T024533Z
LOCATION:3006\, Level 3
DTSTART;TZID=America/Los_Angeles:20250623T104500
DTEND;TZID=America/Los_Angeles:20250623T110000
UID:dac_DAC 2025_sess139_RESEARCH2341@linklings.com
SUMMARY:Graph-Guided Transfer Learning to Boost the Efficiency of System-L
 evel Optimization of Analog/Mixed-Signal Circuits
DESCRIPTION:Jiaqi Wang and Georges Gielen (KU Leuven)\n\nThis paper introd
 uces a novel graph-guided transfer learning approach to boost the efficien
 cy of system-level optimization of analog/mixed-signal circuits. The syste
 m-level optimization is based on Reinforcement Learning (RL) in combinatio
 n with Graph Attention Networks (GAT). The results surpass state-of-the-ar
 t in efficiency and optimality. The key innovation is a graph similarity d
 etection method that leverages embedded design knowledge to identify elect
 rical similarities and trade-offs, enhancing knowledge transferability, ev
 en between dissimilar circuit architectures. Applied to the case study of 
 4th-order continuous-time Delta-Sigma analog-to-digital converters, the gr
 aph-based transfer learning framework enhances the RL sampling efficiency,
  reducing the amount of simulations by up to 11x, and improves the optimiz
 ation results by 12.4% compared to optimization from scratch. As the frame
 work accelerates knowledge transfer across different architectures, it can
  boost the optimization efficiency and improve the performance towards a b
 road range of analog/mixed-signal systems.\n\nTopics: EDA\n\nTracks: EDA6:
  Analog CAD, Simulation, Verification and Test\n\nSession Chairs: Arindam 
 Basu (School of Computer Science and Engineering, Nanyang Technological Un
 iversity) and Markus Olbrich (Leibniz University Hannover)\n\n
END:VEVENT
END:VCALENDAR