BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:Europe/Rome 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:20260717T081801Z UID:A115CE8C-A1F5-4CC6-B9C9-BFD5DB0D09A0 DTSTART;TZID=Europe/Rome:20260720T100000 DTEND;TZID=Europe/Rome:20260720T110000 DESCRIPTION:IEEE ComSoc Italy Chapter Distinguished Lecture Tour\n\nSpeaker : Prof. Carlo Fischione (KTH\, IEEE ComSoc Distinguished Lecturer)\n\nTitl e of Talk: Efficient Federated Learning over Networks\n\nAbstract: Federat ed learning is a distributed stochastic optimization problem in which mult iple agents collaboratively minimize a global objective through local comp utations and iterative information exchange. While classical distributed o ptimization theory often assumes reliable and ideal communication links\, wireless systems introduce severe constraints in terms of bandwidth\, late ncy\, interference\, fading\, and energy consumption. As a result\, commun ication becomes a fundamental bottleneck for scalability and convergence. This seminar presents recent advances on communication computation efficie nt federated learning over networks\, with emphasis on the interplay betwe en optimization theory\, information transmission\, and function computati on. The talk discusses distributed stochastic optimization methods under c ommunication constraints\, including compressed gradient methods\, local-u pdate schemes\, adaptive aggregation\, and convergence analysis under impe rfect information exchange. Finally\, the seminar discusses emerging theor etical directions toward Networked AI\, including learning under communica tion uncertainty\, non-coherent distributed computation\, reliability-awar e learning\, and the joint design of communication\, computation\, and inf erence in large-scale distributed AI systems.\n\nCo-sponsored by: Gran Sas so Science Institute\n\nSpeaker(s): Carlo Fischione\n\nGran Sassos Science Institute\, L'Aquila\, Abruzzi\, Italy LOCATION:Gran Sassos Science Institute\, L'Aquila\, Abruzzi\, Italy ORGANIZER:daniele.tarchi@unifi.it SEQUENCE:12 SUMMARY:Efficient Federated Learning over Networks URL;VALUE=URI:https://events.vtools.ieee.org/m/568363 X-ALT-DESC:Description: <br /><p><strong>IEEE ComSoc Italy Chapter Distingu ished Lecture Tour</strong></p>\n<p><strong>Speaker:</strong> Prof. Carlo Fischione (KTH\, IEEE ComSoc Distinguished Lecturer)</p>\n<p><strong>Title of Talk:</strong> Efficient Federated Learning over Networks</p>\n<p><str ong>Abstract:</strong> Federated learning is a distributed stochastic opti mization problem in which multiple agents collaboratively minimize a globa l objective through local computations and iterative information exchange. While classical distributed optimization theory often assumes reliable an d ideal communication links\, wireless systems introduce severe constraint s in terms of bandwidth\, latency\, interference\, fading\, and energy con sumption. As a result\, communication becomes a fundamental bottleneck for scalability and convergence. This seminar presents recent advances on com munication computation efficient federated learning over networks\, with e mphasis on the interplay between optimization theory\, information transmi ssion\, and function computation. The talk discusses distributed stochasti c optimization methods under communication constraints\, including compres sed gradient methods\, local-update schemes\, adaptive aggregation\, and c onvergence analysis under imperfect information exchange. Finally\, the se minar discusses emerging theoretical directions toward Networked AI\, incl uding learning under communication uncertainty\, non-coherent distributed computation\, reliability-aware learning\, and the joint design of communi cation\, computation\, and inference in large-scale distributed AI systems .</p> END:VEVENT END:VCALENDAR