BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:Europe/Vienna BEGIN:DAYLIGHT DTSTART:20260329T030000 TZOFFSETFROM:+0100 TZOFFSETTO:+0200 RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3 TZNAME:CEST END:DAYLIGHT BEGIN:STANDARD DTSTART:20261025T020000 TZOFFSETFROM:+0200 TZOFFSETTO:+0100 RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10 TZNAME:CET END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260402T123216Z UID:DC353325-2613-400E-B3C5-41EFC2741511 DTSTART;TZID=Europe/Vienna:20260407T150000 DTEND;TZID=Europe/Vienna:20260407T160000 DESCRIPTION:Talk of Prof. Kaushik Sengupta\, Princeton University\n\nAbstra ct: Traditionally\, chip-scale RF system design has been in the domain of the expert\, dominated by thumb rules and trial and error techniques. Desi gning these ICs\, that form the bedrock of the wireless networks\, is comp lex\, time-consuming\, requires years of expertise\, and therefore\, can b e very expensive. Historically\, the design process of RF IC design has re lied on intuition based approaches with standard templates that are subseq uently optimized\, time-consuming parameter sweeps\, or ad-hoc population- based metaheuristic optimization methods. There is no reason to believe th at this approach is optimal in any sense. This talk will discuss how inver se design with AI-based approaches can open a new design space and allow r apid designs on demand. It will discuss deep-learning based modeling and g enerative AI approaches\, that are transferable across process design tech nologies\, for inverse design and automated synthesis of mmWave/sub-THz ci rcuits and antennas.\n\nCo-sponsored by: JKU/SAL Intelligent Wireless Syst ems Lab\n\nSpeaker(s): Kaushik Sengupta\n\nRoom: SAL meeting room\, 4th fl oor\, Bldg: Science Park 4\, Johannes Kepler University Linz\, Altenberger str. 69\, Linz\, Oberosterreich\, Austria\, 4040 LOCATION:Room: SAL meeting room\, 4th floor\, Bldg: Science Park 4\, Johann es Kepler University Linz\, Altenbergerstr. 69\, Linz\, Oberosterreich\, A ustria\, 4040 ORGANIZER:harald.pretl@jku.at SEQUENCE:23 SUMMARY:Talk@JKU: AI-enabled Chip (RF) Design beyond Human Intuition\, by P rof. Sengupta\, Princeton'U URL;VALUE=URI:https://events.vtools.ieee.org/m/552605 X-ALT-DESC:Description: <br /><p>Talk of <span class="fontstyle0"><strong>P rof. Kaushik Sengupta</strong>\, Princeton University</span></p>\n<p><stro ng>Abstract:&nbsp\;</strong><span class="fontstyle0">Traditionally\, chip- scale RF system design has been in the domain of the expert\, </span><span class="fontstyle0">dominated by thumb rules and trial and error technique s. Designing these ICs\, </span><span class="fontstyle0">that form the bed rock of the wireless networks\, is complex\, time-consuming\, </span><span class="fontstyle0">requires years of expertise\, and therefore\, can be v ery expensive. Historically\, </span><span class="fontstyle0">the design p rocess of RF IC design has relied on intuition based approaches with </spa n><span class="fontstyle0">standard templates that are subsequently optimi zed\, time-consuming parameter&nbsp\;</span><span class="fontstyle0">sweep s\, or ad-hoc population-based metaheuristic optimization methods. There < /span><span class="fontstyle0">is no reason to believe that this approach is optimal in any sense. This talk will </span><span class="fontstyle0">di scuss how inverse design with AI-based approaches can open a new design </ span><span class="fontstyle0">space and allow rapid designs on demand. It will discuss deep-learning based </span><span class="fontstyle0">modeling and generative AI approaches\, that are transferable across process </span ><span class="fontstyle0">design technologies\, for inverse design and aut omated synthesis of mmWave/</span><span class="fontstyle0">sub-THz circuit s and antennas.</span>&nbsp\;</p> END:VEVENT END:VCALENDAR