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:3000\, Level 3
DTSTART;TZID=America/Los_Angeles:20250625T113000
DTEND;TZID=America/Los_Angeles:20250625T114500
UID:dac_DAC 2025_sess106_RESEARCH297@linklings.com
SUMMARY:Self-Attention To Operator Learning-based 3D-IC Thermal Simulation
DESCRIPTION:Zhen Huang and hong wang (University of Science and Technology
  of China); Wenkai Yang (ShanghaiTech University); Muxi Tang (Tsinghua Uni
 versity); Depeng Xie (BTD Technology, Inc); Ting-Jung Lin (Eastern Institu
 te of Technology); Yu Zhang (University of Science and Technology of China
 ); Wei Xing (University of Sheffield); and Lei He (Eastern Institute of Te
 chnology)\n\nThermal management in 3D ICs faces challenges due to higher p
 ower densities. Traditional PDE-based simulation methods are accurate but 
 too slow for iterative design. Machine learning methods like FNO offer alt
 ernatives but struggle with high-frequency information loss and reliance o
 n high-fidelity training data. We propose SAU-FNO, integrating self-attent
 ion and U-Net with FNO to capture long-range dependencies and local featur
 es, enhancing high-frequency modeling. Transfer learning further reduces h
 igh-fidelity data needs and accelerates training. SAU-FNO achieves state-o
 f-the-art thermal prediction and an 842× speedup over conventional FEM met
 hods, offering an efficient solution for advanced 3D IC thermal management
 .\n\nTopics: AI\n\nTracks: AI2: AI/ML Application and Infrastructure\n\nSe
 ssion Chairs: Jun Shiomi (The University of Osaka) and Subhajit Dutta Chow
 dhury (Advanced Micro Devices (AMD))\n\n
END:VEVENT
END:VCALENDAR