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:20260402T024534Z
LOCATION:Level 2 Lobby
DTSTART;TZID=America/Los_Angeles:20250623T180000
DTEND;TZID=America/Los_Angeles:20250623T190000
UID:dac_DAC 2025_sess287_LBR108@linklings.com
SUMMARY:Late Breaking Results: Automated Topology Generation for Power Amp
 lifier Designs through BiLSTM-based DNN and Multi-objective Optimizations
DESCRIPTION:Lida Kouhalvandi (Dogus University); Sercan Aygun (University 
 of Louisiana, Lafayette); M. Hassan Najafi (Case Western Reserve Universit
 y); and Arman Roohi (University of Illinois, Chicago)\n\nThis work present
 s an automated methodology for optimizing power amplifier (PA) design by p
 redicting the most suitable circuit topology. Bidirectional long short-ter
 m memory (BiLSTM) deep neural network (DNN) is trained to determine the op
 timal PA topology, while multi-objective Pareto front optimization refines
  the network hyperparameters. The proposed approach is validated through h
 igh-performance PAs using lumped elements and transmission lines at a 1–2 
 GHz frequency range. The method is demonstrated using the Cree CGH40010 Ga
 N HEMT on a Rogers RO4350B substrate, achieving power output of ∼40 dBm, p
 ower-added efficiency of at least 50%, and power gain exceeding 10 dB.\n\n
END:VEVENT
END:VCALENDAR