BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/Los_Angeles BEGIN:DAYLIGHT DTSTART:20250309T030000 TZOFFSETFROM:-0800 TZOFFSETTO:-0700 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:PDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20241103T010000 TZOFFSETFROM:-0700 TZOFFSETTO:-0800 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:PST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20250129T055510Z UID:82BBCF84-A997-4895-9911-D0A7D5264A1E DTSTART;TZID=America/Los_Angeles:20250128T160000 DTEND;TZID=America/Los_Angeles:20250128T180000 DESCRIPTION:Fog learning is an emerging paradigm for optimizing the orchest ration of artificial intelligence services over contemporary network syste ms. Different from existing distributed techniques such as federated learn ing\, fog learning emphasizes intrinsically in its design the unique node\ , network\, and data properties encountered in today's fog networks that s pan computing elements from the edge to the cloud. An important thread of research in fog learning has been on understanding the role that local top ologies formed on an ad-hoc basis among proximal groups of heterogeneous c omputing elements can play in elevating the achievable tradeoff between in telligence quality and resource efficiency. In this talk\, I will discuss recent results on the analysis of fog learning processes which give insigh ts into the impact that these topologies\, along with other properties suc h as model characteristics and fog decision parameters\, have on global tr aining performance. Additionally\, I will discuss the development of adapt ive control methodologies that leverage such relationships for jointly opt imizing relevant fog learning metrics.\n\nSpeaker(s): Dr. Christopher Brin ton\n\nAgenda: \nNetworking: 4:00 PM to 4:30 PM (Refreshments will be prov ided)\n\nPresentation: 4:30 PM to 5:15 PM\n\nQ&A: 5:15 PM to 5:45 PM\n\nNe tworking: 5:45 PM to 6:00 PM\n\nRoom: 206\, Bldg: Daly Science Center\, Sa nta Clara University\, Santa Clara\, California\, United States\, 95050 LOCATION:Room: 206\, Bldg: Daly Science Center\, Santa Clara University\, S anta Clara\, California\, United States\, 95050 ORGANIZER:bdezfouli@scu.edu SEQUENCE:7 SUMMARY:FROM FEDERATED TO FOG LEARNING: EXPANDING THE FRONTIER OF MODEL TRA INING OVER CONTEMPORARY WIRELESS NETWORK SYSTEMS URL;VALUE=URI:https://events.vtools.ieee.org/m/463566 X-ALT-DESC:Description: <br /><p>Fog learning is an emerging paradigm for o ptimizing the orchestration of artificial intelligence services over conte mporary network systems. Different from existing distributed techniques su ch as federated learning\, fog learning emphasizes intrinsically in its de sign the unique node\, network\, and data properties encountered in today' s fog networks that span computing elements from the edge to the cloud. An important thread of research in fog learning has been on understanding th e role that local topologies formed on an ad-hoc basis among proximal grou ps of heterogeneous computing elements can play in elevating the achievabl e tradeoff between intelligence quality and resource efficiency. In this t alk\, I will discuss recent results on the analysis of fog learning proces ses which give insights into the impact that these topologies\, along with other properties such as model characteristics and fog decision parameter s\, have on global training performance. Additionally\, I will discuss the development of adaptive control methodologies that leverage such relation ships for jointly optimizing relevant fog learning metrics.</p>\n<p>&nbsp\ ;</p><br /><br />Agenda: <br /><p>Networking: 4:00 PM to 4:30 PM (Refreshm ents will be provided)</p>\n<p>Presentation: 4:30 PM to 5:15 PM</p>\n<p>Q& amp\;A: 5:15 PM to 5:45 PM</p>\n<p>Networking: 5:45 PM to 6:00 PM</p>\n<p> &nbsp\;</p> END:VEVENT END:VCALENDAR