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:3004\, Level 3
DTSTART;TZID=America/Los_Angeles:20250624T141500
DTEND;TZID=America/Los_Angeles:20250624T143000
UID:dac_DAC 2025_sess142_RESEARCH1208@linklings.com
SUMMARY:Reinforcement Learning-Driven Window Selection for Enhanced Window
 -Based Rip-up and Reroute in Chip Detailed Routing
DESCRIPTION:Yu-Chan Keng and Yu-Chun Pai (National Yang Ming Chiao Tung Un
 iversity); Wen-Hao Liu, Haoxing Ren, Danny Liu, Rongjian Liang, Mark Ho, a
 nd Anthony Agnesina (Nvidia); and Yih-Lang Li (National Yang Ming Chiao Tu
 ng University)\n\nWith increasingly complex design rules and pin density i
 n advanced technology nodes, achieving a violation-free layout has become 
 more challenging, also making rip-up and reroute (RUR) the most runtime-in
 tensive component of detailed routing. We propose a novel reinforcement le
 arning (RL)-based approach to enhance the window-based RUR process. Our me
 thod features a dynamic window generation strategy that adjusts window siz
 e and position based on the distribution of design rule violations (DRV), 
 enabling efficient targeting of congested areas. By leveraging the predict
 ive capabilities of RL, our approach aims to minimize DRVs and achieve hig
 h-quality routing results. Experimental results demonstrate that our metho
 d outperforms the state-of-the-art detailed routers, TritonRoute, achievin
 g a DRV-free solution, averagely improving wirelength by 0.07%, via count 
 by 2.42%, and consuming almost the same average runtime.\n\nTopics: EDA\n\
 nTracks: EDA7: Physical Design and Verification\n\nSession Chairs: Michael
  Kazda (IBM) and Stephan Held (University of Bonn)\n\n
END:VEVENT
END:VCALENDAR