BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/New_York BEGIN:DAYLIGHT DTSTART:20260308T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20261101T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20261005T132531Z UID:DFE928D2-5C59-4FFF-B908-C266210EE5FD DTSTART;TZID=America/New_York:20260930T180000 DTEND;TZID=America/New_York:20260930T193000 DESCRIPTION:AI and machine learning are unconventional technologies with un ique capability to address challenges in electromagnetic-based analysis an d optimization in high-speed/high-frequency electronic components/packages and systems. With phenomenal progress in electromagnetic based computatio n algorithms\, along with dramatic changes in the computing environment\, high-fidelity electromagnetic models are now an important part of high-spe ed/high-frequency electromagnetic design automation. However new design ch allenges continue to rise. Electromagnetic structures and circuits are bec oming more complex\, and frequency is getting higher. More sophistication in multiphysics modeling and design are becoming increasingly necessary. M eaningful design problems easily become computationally prohibitive.\n\nIn this talk\, we present AI/machine learning technologies for electromagnet ic/multiphysics based modeling and optimization\, and their applications t o signal/power integrity analysis of highspeed/high-frequency electronic p ackages and subsystems. We will highlight emerging directions of knowledge -based\, cognition-driven design. Incorporating domain-specific design kno wledge/engineering equations into artificial neural networks\, knowledge-b ased and deep-learning based computational technologies are producing fine -grained modeling and design solutions for problems which are otherwise co mputationally very expensive. New formulations of inverse neural network t raining algorithms allow instant solutions to electromagnetic inverse mode ling problems addressing the technical challenges of non-uniqueness in inv erse modeling. Emerging machine learning structures and optimization algor ithms for electromagnetic based design and signal/power integrity analysis will be discussed.\n\nThis talk is open to everyone\, IEEE membership is not required.\n\nPlease join us at 5:30PM for networking and light refresh ments!\n\nCo-sponsored by: Syracuse University EECS Department\n\nSpeaker( s): IEEE EMC-S Distinguished Lecturer and IEEE Fellow Prof. Qi-jun Zhang\n \nAgenda: \n5:30pm: Networking and light refreshments\;\n\n6:00pm: Present ation\;\n\n7:00pm: Q&A and discussion\;\n\nRoom: CST 4-201\, Bldg: Parking : Street\, or paid parking in one of SU's parking lots.\, Center for Scien ce and Technology at Syracuse University\, 111 College Pl\, Syracuse\, New York\, United States\, 13210 LOCATION:Room: CST 4-201\, Bldg: Parking: Street\, or paid parking in one o f SU's parking lots.\, Center for Science and Technology at Syracuse Unive rsity\, 111 College Pl\, Syracuse\, New York\, United States\, 13210 ORGANIZER:menders@ieee.org SEQUENCE:33 SUMMARY:Emerging AI/Machine Learning Technologies for Analysis and Optimiza tion in High-Speed/High-Frequency Packages and Systems URL;VALUE=URI:https://events.vtools.ieee.org/m/578419 X-ALT-DESC:Description: <br /><p>AI and machine learning are unconventional technologies with unique capability to address challenges in electromagne tic-based analysis and optimization in high-speed/high-frequency electroni c components/packages and systems. With phenomenal progress in electromagn etic based computation algorithms\, along with dramatic changes in the com puting environment\, high-fidelity electromagnetic models are now an impor tant part of high-speed/high-frequency electromagnetic design automation. However new design challenges continue to rise. Electromagnetic structures and circuits are becoming more complex\, and frequency is getting higher. More sophistication in multiphysics modeling and design are becoming incr easingly necessary. Meaningful design problems easily become computational ly prohibitive.</p>\n<p>In this talk\, we present AI/machine learning tech nologies for electromagnetic/multiphysics based modeling and optimization\ , and their applications to signal/power integrity analysis of highspeed/h igh-frequency electronic packages and subsystems. We will highlight emergi ng directions of knowledge-based\, cognition-driven design. Incorporating domain-specific design knowledge/engineering equations into artificial neu ral networks\, knowledge-based and deep-learning based computational techn ologies are producing fine-grained modeling and design solutions for probl ems which are otherwise computationally very expensive. New formulations o f inverse neural network training algorithms allow instant solutions to el ectromagnetic inverse modeling problems addressing the technical challenge s of non-uniqueness in inverse modeling. Emerging machine learning structu res and optimization algorithms for electromagnetic based design and signa l/power integrity analysis will be discussed.</p>\n<p>&nbsp\;</p>\n<p><str ong>This talk is open to everyone\, IEEE membership is not required.</stro ng></p>\n<p><strong>Please join us at 5:30PM for networking and light refr eshments!</strong></p>\n<p>&nbsp\;</p><br /><br />Agenda: <br /><p>5:30pm: Networking and light refreshments\;</p>\n<p><strong>6:00pm: Presentation\ ;</strong></p>\n<p>7:00pm: Q&amp\;A and discussion\;</p> END:VEVENT END:VCALENDAR