BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//TECoSA - ECPv6.17.3.1//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:TECoSA
X-ORIGINAL-URL:https://www.tecosa.center.kth.se
X-WR-CALDESC:Events for TECoSA
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:UTC
BEGIN:STANDARD
TZOFFSETFROM:+0000
TZOFFSETTO:+0000
TZNAME:UTC
DTSTART:20240101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;VALUE=DATE:20250901
DTEND;VALUE=DATE:20250905
DTSTAMP:20250923T140907Z
CREATED:20250923T140855Z
LAST-MODIFIED:20250923T140907Z
UID:7788-1756684800-1757030399@www.tecosa.center.kth.se
SUMMARY:Edge AI Summer School
DESCRIPTION:Two weeks ago\, we wrapped up the Summer School on Edge Artificial Intelligence (1–4 Sept 2025) at KTH Royal Institute of Technology\, Stockholm — an exciting four-day journey into the cutting edge of Edge AI\, co-organized by Prof. James Gross and Adarsh Behera. With over 50 participants and speakers from academia and industry across 20+ institutions\, the program featured keynotes\, research talks\, tutorials\, and vibrant discussions. We’re incredibly grateful to everyone who made this a success! \nA huge thank you to our speakers for their thought-provoking contributions: \n\n Eyal de Lara (University of Toronto) kicked things off with insights into edge system design — orchestration\, component boundaries\, and edge preprocessing to reduce latency\, bandwidth\, and improve privacy.\nJaya Prakash Champati (University of Victoria) introduced hierarchical inference for LLM routing and QoS-aware edge scheduling under dynamic workloads.\nHenning Schulzrinne (Columbia University) asked: “Edge AI – Always just beyond the horizon?” He brought a real-world lens to Edge AI in telecom and automotive.\nXiang Su (University of Helsinki) explored model/system heterogeneity in federated learning and the trade-offs of personalization vs. global efficiency.\nQi Zhang (Aarhus University) discussed semantic communication and how goal-oriented transmission can make Edge AI more context-aware and efficient.\nRoberto Morabito (EURECOM) gave a future-looking talk on TinyML and Generative AI at the Edge\, pushing the boundary of edge-to-device intelligence.\nRafia Inam (Ericsson/KTH Royal Institute of Technology) highlighted the role of Trustworthy & Explainable AI in telecom\, ongoing EU standardization\, and real-world XAI limitations.\nElla Peltonen (University of Oulu) showcased practical Edge AI use cases in the automotive sector — with insights into deployment and challenges.\n\nA big thank you to our sponsors: NordForsk\, Digital Futures\, TECoSA – Center for Trustworthy Edge Computing Systems and Applications\, and the Nordic University Cooperation on Edge Intelligence for their generous support. \nAnd of course\, a big thanks to all the participants — your energy\, curiosity\, and collaboration brought the event to life!
URL:https://www.tecosa.center.kth.se/event/edge-ai-summer-school/
CATEGORIES:Conference,Workshops
END:VEVENT
END:VCALENDAR