BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Department of Electrical and Computer Engineering (HKUECE) 電機與計算機工程系 - ECPv6.16.0//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-ORIGINAL-URL:https://ece.hku.hk
X-WR-CALDESC:Events for Department of Electrical and Computer Engineering (HKUECE) 電機與計算機工程系
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:Asia/Hong_Kong
BEGIN:STANDARD
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
TZNAME:HKT
DTSTART:20220101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20231221T110000
DTEND;TZID=Asia/Hong_Kong:20231221T120000
DTSTAMP:20260513T073656
CREATED:20231213T085231Z
LAST-MODIFIED:20250114T075234Z
UID:17882-1703156400-1703160000@ece.hku.hk
SUMMARY:RPG Seminar – Modelling AC Loss in Superconductors via Integral Method
DESCRIPTION:Superconductors can potentially be used in electrical machines in future electric aircraft\, since superconductors can increase the power density of electrical machines. When superconductors are carrying ac or are subject to an alternating magnetic field\, they experience ac loss. AC loss affects the efficiency of the machines and the cooling power needed\, which has implications for the mass of the overall system of the electric aircraft. The integral method can model superconductors that are carrying arbitrary ac and under an arbitrary external magnetic field. This talk will review the integral method in the literature\, and explain how the integral method can be used to model superconductors in an electrical machine. In addition\, it will also explain how the integral method can be used to model in 2D cables made of superconducting tapes that are coupled (electrically connected at the ends of the cable or along the whole length of the tapes). \nZoom Link :\nhttps://hku.zoom.us/j/99033734395\nMeeting ID: 990 3373 4395 \nBiography of the speaker:\n\nChung Tin Calvin Chow received the B.A. and M.Eng. degrees in engineering from the University of Cambridge\, Cambridge\, U.K.\, both in 2020\, with specialization in areas including control and information engineering. He is currently pursuing a Ph.D. degree in electrical and electronic engineering with the Department of Electrical and Electronic Engineering\, The University of Hong Kong\, Hong Kong\, SAR\, China. He was a visitor at Karlsruhe Institute of Technology\, Germany\, for around 5 months in 2022-2023 and at the University of Strathclyde\, UK\, for around 6 months in 2023. His research interests include superconducting machines and drives\, superconductor modelling and experimentation. \nAll are welcome.
URL:https://ece.hku.hk/events/rpg-seminar-modelling-ac-loss-in-superconductors-via-integral-method/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20231221T140000
DTEND;TZID=Asia/Hong_Kong:20231221T150000
DTSTAMP:20260513T073656
CREATED:20231214T090442Z
LAST-MODIFIED:20250114T075128Z
UID:17883-1703167200-1703170800@ece.hku.hk
SUMMARY:RPG Seminar – Internal Cross-layer Gradients for Extending Homogeneity to Heterogeneity in Federated Learning
DESCRIPTION:Federated learning (FL) inevitably confronts the challenge of system heterogeneity in practical scenarios. To enhance the capabilities of most model-homogeneous FL methods in handling system heterogeneity\, we propose a training scheme that can extend their capabilities to cope with this challenge. In this seminar\, we commence our study with a detailed exploration of homogeneous and heterogeneous FL settings and discover three key observations: (1) a positive correlation between client performance and layer similarities\, (2) higher similarities in the shallow layers in contrast to the deep layers\, and (3) the smoother gradient distributions indicate the higher layer similarities. Building upon these observations\, we introduce InCo Aggregation that leverages internal cross-layer gradients\, a mixture of gradients from shallow and deep layers within a server model\, to augment the similarity in the deep layers without requiring additional communication between clients. Furthermore\, our methods can be tailored to accommodate model-homogeneous FL methods such as FedAvg\, FedProx\, FedNova\, Scaffold\, and MOON\, to expand their capabilities to handle the system heterogeneity. Copious experimental results validate the effectiveness of InCo Aggregation\, spotlighting internal cross-layer gradients as a promising avenue to enhance the performance in heterogeneous FL. \nZoom Link :\nhttps://hku.zoom.us/j/93707944044?pwd=VEtnNkVzYnNlY2IrUWR0UjhwVGV5UT09 \nBiography of the speaker:\n\nYun-Hin Chan received his B.Eng. of Software Engineering from Sun Yat-sen University. Currently\, he is pursuing his Ph.D. degree at the University of Hong Kong. His work is focused on how to solve practical challenges in federated learning\, such as communication efficiency and system heterogeneity. His research interests include deep learning\, distributed optimization\, federated learning\, knowledge distillation\, and transfer learning. \nAll are welcome.
URL:https://ece.hku.hk/events/rpg-seminar-internal-cross-layer-gradients-for-extending-homogeneity-to-heterogeneity-in-federated-learning/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
END:VCALENDAR