BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Department of Electrical and Computer Engineering (HKUECE) 電機與計算機工程系 - ECPv6.15.20//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:20240101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20250506T100000
DTEND;TZID=Asia/Hong_Kong:20250506T110000
DTSTAMP:20260509T183754
CREATED:20250603T031739Z
LAST-MODIFIED:20250626T094107Z
UID:111541-1746525600-1746529200@ece.hku.hk
SUMMARY:A Diamond Heater-Thermometer Microsensor for Measuring Localized Thermal Conductivity: A Case Study in Gelatin Hydrogel
DESCRIPTION:Zoom Link: https://hku.zoom.us/j/99573774863?pwd=rkA2rKib5AXIzhfMD5grx9darAa05B.1\nMeeting ID: 995 7377 4863\nPassword: 437351 \nAbstract\nUnderstanding the microscopic thermal effects of the hydrogel is important for its application in diverse fields\, including thermal-related studies in tissue engineering and thermal management for flexible electronic devices. In recent decades\, localized thermal properties\, such as thermal conductivity\, have often been overlooked due to technical limitations. To tackle this\, the study proposes a new hybrid diamond microsensor that is capable of simultaneous temperature control and readout in a decoupled manner. Specifically\, the sensor consists of a silicon pillar (heater) at ≈10 microns in length\, topped by a micron-sized diamond particle that contains silicon-vacancy (SiV) centers (thermometer) with 1.29 K/Hz^−0.5 temperature measurement sensitivity. Combining this innovative\, scalable sensor with a newly established simulation model that can transform heating-laser-induced temperature change into thermal conductivity\, an all-optical decoupled method is introduced with ≈0.05 W m−1 K−1 precision\, which can reduce laser crosstalk. For the first time\, the thermal conductivity change of hydrogels during the gelation process is tracked and the existence of variation is demonstrated. The study introduces a rapid\, undisturbed technique for measuring microscale thermal conductivity\, potentially serving as a valuable tool for cellular thermometry\, and highlights the idea that decoupling can reduce crosstalk from different lasers\, which is helpful for quantum sensing. \nSpeaker\nMr. Ma Linjie\nDepartment of Electrical and Electronic Engineering\nThe University of Hong Kong \nSpeaker’s Biography\nLinjie Ma is currently pursuing a Ph.D. degree in the Department of Electrical and Electronic Engineering at The University of Hong Kong\, under the supervision of Prof. Zhiqin Chu. He received his B.S. degree in Physics from Nanjing University (NJU)\, China\, in 2020. His research focuses on nitrogen-vacancy (NV) centers\, diamond-based biosensing\, and mechanobiology. \nAll are welcome!
URL:https://ece.hku.hk/events/20250506-0/
LOCATION:Online via Zoom
CATEGORIES:Highlights,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:20250506T110000
DTEND;TZID=Asia/Hong_Kong:20250506T120000
DTSTAMP:20260509T183754
CREATED:20250603T025307Z
LAST-MODIFIED:20250603T025832Z
UID:111497-1746529200-1746532800@ece.hku.hk
SUMMARY:Progressive End-to-End Object Detection in Crowded Scenes
DESCRIPTION:Zoom Link: https://hku.zoom.us/j/92481644796?pwd=2CJcxbzAimj87HfFHoBMvUr9oCnOUZ.1\nAbstract\nCrowded object detection is a practical yet challenging research field in computer vision. Many research efforts have been made and achieved impressive progress in the last few decades. However\, most of them require handcraft components\, e.g. anchor settings and post-processing\, resulted in sub-optimal performance in handling scenes. In this work\, we propose a new query-based detection framework for crowd detection. Previous query-based detectors suffer from two drawbacks: first\, multiple predictions will be inferred for a single object\, typically in crowded scenes; second\, the performance saturates as the depth of the decoding stage increases. Benefiting from the nature of the one-to-one label assignment rule\, we propose a progressive predicting method to address the above issues. Specifically\, we first select accepted queries prone to generate true positive predictions\, then refine the rest noisy queries according to the previously accepted predictions. Experiments show that our method can significantly boost the performance of query-based detectors in crowded scenes. Moreover\, the proposed method\, robust to crowdedness\, can still obtain consistent improvements on moderately and slightly crowded datasets\, such as CityPersons and COCO. \nSpeaker\nMr. Zheng Anlin\nDepartment of Electrical and Electronic Engineering\nThe University of Hong Kong \nSpeaker’s Biography\nMr. Anlin Zheng received the M.S. degree in computer science and technology from Beihang University\, China. He then joined Beijing Megvii Technology Co.\, Ltd. He is currently pursuing the Ph.D. degree in electrical and electronic engineering from the University of Hong Kong (HKU)\, under the supervision of Dr. Xiaojuan Qi. His research focuses on applying deep learning technology to computer vision\, including object detection and AIGC. \nAll are welcome!
URL:https://ece.hku.hk/events/20250506-3/
LOCATION:Online via Zoom
CATEGORIES:Highlights,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:20250506T140000
DTEND;TZID=Asia/Hong_Kong:20250506T150000
DTSTAMP:20260509T183754
CREATED:20250603T023635Z
LAST-MODIFIED:20250603T023709Z
UID:111455-1746540000-1746543600@ece.hku.hk
SUMMARY:Highly Integrated Wireless Direct Drive Motor System for Fully Enclosed Environments
DESCRIPTION:Zoom Link: https://hku.zoom.us/j/95286760379?pwd=3C7A8QPlmXLJAQZ5IsYbEWdDXXvWk1.1\nMeeting ID: 952 8676 0379\nPassword: 492840\nAbstract\nContemporary global motor systems predominantly rely on cable-based and battery-powered energy transmission mediums\, both of which exhibit fundamental structural limitations. Wired systems face challenges such as installation and maintenance complexity\, mobility constraints\, and safety vulnerabilities\, particularly in confined spaces or high-precision applications. On the other hand\, battery-dependent systems struggle with energy density limitations\, thermal sensitivity\, and a mass penalty\, which can increase operational downtime and affect performance. These systemic deficiencies underscore the urgent need to develop next-generation motor systems that integrate contactless power transfer technologies\, wireless direct-drive control\, and passive intelligent control to overcome conventional electromechanical constraints. This study introduces a highly integrated wireless ultrasonic motor system featuring three fundamental innovations. First\, an integrated magnetic coupler is designed to realize independent decoupling control of two-phase high-frequency magnetic fields\, eliminating the dependence on cables and batteries at the receiving side. In addition\, the structural design of the receiving side is simplified to realize the synchronous transmission of wireless energy and wireless drive signals\, enabling the high-frequency electromagnetic energy (40 kHz) induced at the receiving side can directly drive the motor\, which breaks through the elimination of the rectifier and inverter link. Furthermore\, an intelligent passive control is proposed\, whereby the motor side realizes the complete elimination of components such as controllers\, sensors\, compensation capacitors\, semiconductor switches\, rectifiers\, etc.\, and the precise control of rotational speed and direction can be accomplished through the electromagnetic field modulation at the transmitter side\, resulting in a significant reduction of system complexity and cost.\nSpeaker\nMr. Zhiwei Xue\nDepartment of Electrical and Electronic Engineering\nThe University of Hong Kong\nSpeaker’s Biography\nZhiwei Xue is currently working toward the Ph.D. degree in electrical and electronic engineering with the Department of Electrical and Electronic Engineering at the University of Hong Kong\, Hong Kong\, China. From 2021 to 2022\, he was a Research Assistant at the Department of Electrical Engineering\, The Hong Kong Polytechnic University\, Hong Kong\, China. His research interests include wireless power transfer\, electrical machine drives\, and power electronics. \nAll are welcome!
URL:https://ece.hku.hk/events/20250506-2/
LOCATION:Online via Zoom
CATEGORIES:Highlights,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:20250507T100000
DTEND;TZID=Asia/Hong_Kong:20250507T110000
DTSTAMP:20260509T183754
CREATED:20250603T025022Z
LAST-MODIFIED:20250603T025022Z
UID:111484-1746612000-1746615600@ece.hku.hk
SUMMARY:Towards Ubiquitous Radio Access Using Nanodiamond Based Quantum Receivers
DESCRIPTION:Zoom Link: https://hku.zoom.us/j/5098778281?pwd=wMZ3GQvpRdxkCjv8p79h3JN1xdgOJe.1\nMeeting ID: 509 877 8281\nPassword: 670951\nAbstract\nThe development of sixth-generation wireless communication systems demands innovative solutions to address challenges in the deployment of a large number of base stations and the detection of multi-band signals. Quantum technology\, specifically nitrogen-vacancy centers in diamonds\, offers promising potential for the development of compact\, robust receivers capable of supporting multiple users. Here we propose a multiple access scheme using fluorescent nanodiamonds containing nitrogen-vacancy centers as nano-antennas. The unique response of each nanodiamond to applied microwaves allows for distinguishable patterns of fluorescence intensities\, enabling multi-user signal demodulation. We demonstrate the effectiveness of our nanodiamonds-implemented receiver by simultaneously transmitting two uncoded digitally modulated information bit streams from two separate transmitters\, achieving a low bit error ratio. Moreover\, our design supports tunable frequency band communication and reference-free signal decoupling\, reducing communication overhead. Furthermore\, we implement a miniaturized device comprising all essential components\, highlighting its practicality as a receiver serving multiple users simultaneously. This approach enables the integration of quantum sensing technologies into future wireless communication networks.\nSpeaker\nMr. Zhang Jiahua\nDepartment of Electrical and Electronic Engineering\nThe University of Hong Kong\nSpeaker’s Biography\n\nJiahua Zhang is currently pursuing a Ph.D. degree in the Department of Electrical and Electronic Engineering at The University of Hong Kong\, under the supervision of Prof. Zhiqin Chu. He received his B.Eng. and M.Eng. degree in Optical Engineering from Harbin Institute of Technology (HIT)\, China\, in 2019 and 2021. His research focuses on nitrogen-vacancy (NV) centers\, diamond-based biosensing\, and thermometry.\n\nAll are welcome!
URL:https://ece.hku.hk/events/20250507-2/
LOCATION:Online via Zoom
CATEGORIES:Highlights,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:20250507T140000
DTEND;TZID=Asia/Hong_Kong:20250507T150000
DTSTAMP:20260509T183754
CREATED:20250603T024130Z
LAST-MODIFIED:20250603T024304Z
UID:111460-1746626400-1746630000@ece.hku.hk
SUMMARY:A Novel Training Framework for Physics-informed Neural Networks: Towards Real-time Applications in Ultrafast Ultrasound Blood Flow Imaging
DESCRIPTION:Abstract\nUltrafast ultrasound blood flow imaging is a state-of-the-art technique for depiction of complex blood flow dynamics in vivo through thousands of full-view image data (or\, timestamps) acquired per second. Physics-informed Neural Network (PINN) is one of the most preeminent solvers of the Navier-Stokes equations\, widely used as the governing equation of blood flow. However\, that current approaches rely on full Navier-Stokes equations is impractical for ultrafast ultrasound. We hereby propose a novel PINN training framework for solving the Navier-Stokes equations. It involves discretizing Navier-Stokes equations into steady state and sequentially solving them with test-time adaptation. The novel training framework is coined as SeqPINN. Upon its success\, we propose a parallel training scheme for all timestamps based on averaged constant stochastic gradient descent as initialization. Uncertainty estimation through Stochastic Weight Averaging Gaussian is then used as an indicator of generalizability of the initialization. This algorithm\, named SP-PINN\, further expedites training of PINN while achieving comparable accuracy with SeqPINN. The performance of SeqPINN and SP-PINN was evaluated through finite-element simulations and in vitro phantoms of single-branch and trifurcate blood vessels. The successful implementation of SeqPINN and SP-PINN open the gate for real-time training of PINN for Navier-Stokes equations and subsequently reliable imaging-based blood flow assessment in clinical practice.\nSpeaker\nMr. Haotian Guan\nDepartment of Electrical and Electronic Engineering\nThe University of Hong Kong\nSpeaker’s Biography\n\nHaotian Guan received his B.S. in Applied Mathematics from The University of New Hampshire in 2019 and the M.S. in Data Science from New York University in 2021. He is currently pursuing the Ph.D. degree in the Department of Electrical and Electronic Engineering at the University of Hong Kong\, Hong Kong.\n\n\nAll are welcome!
URL:https://ece.hku.hk/events/20250507-1/
LOCATION:Room CB-603\, 6/F\, Chow Yei Ching Building\, The University of Hong Kong
CATEGORIES:Highlights,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:20250507T153000
DTEND;TZID=Asia/Hong_Kong:20250507T163000
DTSTAMP:20260509T183754
CREATED:20250603T025756Z
LAST-MODIFIED:20250603T041929Z
UID:111506-1746631800-1746635400@ece.hku.hk
SUMMARY:Rank-Revealing Bayesian Block-Term Tensor Completion with Graph Information
DESCRIPTION:Zoom Link: https://hku.zoom.us/j/97776760951?pwd=zaNBWC786IgVZjQ7NU8SNDJdEeIorn.1 \nAbstract\nBlock-term decomposition (BTD)\, particularly its rank-(L_r\,L_r\,1) special case\, is widely used in signal processing. Traditional methods for computing BTD from fully observed tensors either unrealistically assume the tensor rank and block-term ranks are known or require exhaustive tuning of these parameters. While sparsity-promoting regularization has been introduced to estimate ranks more efficiently\, it still requires regularization parameter tuning. Bayesian learning addresses these issues by employing sparsity-promoting priors\, but so far is limited to fully observed BTD tensors. To process incomplete BTD tensors\, only a few optimization-based methods have been proposed\, and they continue to suffer from time-consuming tuning. To enable tuning-free BTD completion\, a novel prior is proposed here within the Bayesian framework\, and it is proved theoretically that the proposed prior induces the desired dual-level sparsity as well as graph information in the BTD model. A mean-field design is further proposed to develop a closed-form updating variational inference (VI) algorithm without loss of graph information. Extensive experiments on both synthetic datasets and real-world datasets demonstrate the superiority of the proposed method in terms of rank learning\, tensor recovery\, and factor recovery. \nSpeaker\nMr. Zhongtao Chen\nDepartment of Electrical and Electronic Engineering\nThe University of Hong Kong \nSpeaker’s Biography\nZhongtao Chen received the B.Eng. degree from The Chinese University of Hong Kong\, Shenzhen\, China\, in 2021. He is currently working toward the Ph.D. degree with The University of Hong Kong\, Hong Kong. His research interests include signal processing and machine learning using Bayesian methods. \n\n\nAll are welcome!
URL:https://ece.hku.hk/events/20050507-3/
LOCATION:Online via Zoom
CATEGORIES:Highlights,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:20250508T160000
DTEND;TZID=Asia/Hong_Kong:20250508T170000
DTSTAMP:20260509T183754
CREATED:20250603T024218Z
LAST-MODIFIED:20250603T025042Z
UID:111464-1746720000-1746723600@ece.hku.hk
SUMMARY:Trustworthy Image Semantic Communication with GenAI: Explainability\, Controllability\, and Efficiency
DESCRIPTION:Abstract\nImage semantic communication (ISC) has garnered significant attention for its potential to achieve high efficiency in visual content transmission. However\, existing ISC systems based on joint source-channel coding face challenges in interpretability\, operability\, and compatibility. To address these limitations\, we propose a novel trustworthy ISC framework. This approach leverages text extraction and segmentation mapping techniques to convert images into explainable semantics\, while employing Generative Artificial Intelligence (GenAI) for multiple downstream inference tasks. We also introduce a multi-rate ISC transmission protocol that dynamically adapts to both the received explainable semantic content and specific task requirements at the receiver. Simulation results based on a real-world demo demonstrate that our framework achieves explainable learning\, decoupled training\, and compatible transmission in various application scenarios. Finally\, some intriguing research directions and application scenarios are identified. \nSpeaker\nDr. Chenyuan FENG\nMarie Skłodowska-Curie Scholar\,\n6G Star Young Scientist\,\nResearch Fellow at the University of Exeter \nSpeaker’s Biography\nDr. Chenyuan FENG\, Marie Skłodowska-Curie Scholar\, 6G Star Young Scientist. Dr. Feng earned the Ph.D. degree from the Singapore University of Technology and Design. Currently\, Dr. Feng is a Research Fellow at the University of Exeter\, U.K. Her research interests include edge intelligence and AI for communication and network. Dr. Feng has published over 40 papers\, including one ESI top 1% highly cited paper and 3 IEEE conference best papers. Moreover\, Dr. Feng has obtained five Chinese national invention patents and three edited book; earned the First Prize in International Postdoctoral Innovation and Entrepreneurship Competition\, one Gold and one Silver Awards in Chinese Internet+ Innovation and Entrepreneurship Competition; presided one EU horizon project and several National Natural Science Foundation project and national key R&D sub-project\, as well as one Enterprise Start-up Grant for Intelligent Unmanned Systems R & D Project (funded by Merchant & Investment Bureau in Chengdu Government\, China\, 5 million RMB\, as a Co-founder) and one Enterprise Start-up Grant for AI-RAN. She has served as a TPC member in numerous international conferences\, and an Associate Editor for IEEE IoTJ and IEEE OJ-COMS. \nOrganiser\nProf. Hongyang DU\nDepartment of Electrical and Electronic Engineering\,\nThe University of Hong Kong \nAll are welcome!
URL:https://ece.hku.hk/events/20250508-1/
LOCATION:Room CB-601J\, 6/F\, Chow Yei Ching Building\, The University of Hong Kong
CATEGORIES:Highlights,Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2025/06/1280-1.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20250509T150000
DTEND;TZID=Asia/Hong_Kong:20250509T160000
DTSTAMP:20260509T183754
CREATED:20250603T035916Z
LAST-MODIFIED:20250603T035916Z
UID:111587-1746802800-1746806400@ece.hku.hk
SUMMARY:WireLightning: Harnessing Capacitances for In-Transit Massively Parallel Matrix Multiplication
DESCRIPTION:Zoom Link: https://hku.zoom.us/j/3837289217?omn=95617077246 \nAbstract\nAnalog computing-in-memory accelerators promise ultra-low-power\, on-device AI by reducing data transfer and energy usage. Yet inherent device variations and high energy consumption for analog-digital conversion continue to hinder their wide-scale adoption in mainstream systems. To address these issues\, this presentation will introduce WireLightning\, a novel capacitive-computing accelerator featuring a mixed-signal architecture that rethinks analog AI acceleration. Unlike conventional analog crossbars that encode weights in programmable devices\, WireLightning exploits intrinsic charge dynamics in passive capacitors\, encoding matrix multiplication through spike amplitude and timing. This design addresses critical limitations such as weight drift\, stochasticity\, and power-intensive ADC bottlenecks. Key innovations include: amplitude-temporal dual encoding that enables constant-time analog dot-products; time-based decoding scheme that significantly reduces reliance on power-intensive ADCs; row-wise parallel architecture for concurrent dot-product calculations across multiple rows to enhance throughput; and value repetition exploitation in low-bit quantized vectors to reduce multiplications to constant time complexity. A PCB prototype achieved higher accuracies than leading RRAM crossbar and PCM crossbar implementations. Implemented in a 40-nm CMOS technology\, WireLightning demonstrate superior potential in power efficiency\, while maintaining high precision. By integrating algorithm-circuit co-design with physical computing\, this work establishes capacitive computing as a promising path toward combining digital precision and analog efficiency in next-generation edge AI. \nSpeaker\nSpeaker: Mr. WANG Song\nDepartment of Electrical and Electronic Engineering\nThe University of Hong Kong \nSpeaker’s Biography\nSong Wang received the B.Eng. degree in the Department of Automated Test and Control at Harbin Institute of Technology\, and M.Phil. degree in the Department of Mechanical Engineering at the University of Hong Kong. He is currently pursuing the Ph.D. degree in the Department of Electrical and Electronic Engineering at the University of Hong Kong\, under the supervision of Prof. Hayden So. His research interests include AI chip\, computer architecture\, and reconfigurable computing. \n\nAll are welcome!
URL:https://ece.hku.hk/events/20250509-1/
LOCATION:Online via Zoom
CATEGORIES:Highlights,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:20250512T140000
DTEND;TZID=Asia/Hong_Kong:20250512T150000
DTSTAMP:20260509T183754
CREATED:20250603T035657Z
LAST-MODIFIED:20250603T035719Z
UID:111583-1747058400-1747062000@ece.hku.hk
SUMMARY:Towards Federated and Annotation-efficient Deep Learning for Medical Image Analysis
DESCRIPTION:Zoom Link : https://hku.zoom.us/j/91765972342?pwd=CHXoKEknnfPc6zbhHCADi7A1abVUyI.1 \nAbstract\nAs deep learning is increasingly applied in medical image analysis\, developing efficient and accurate models has become crucial. However\, traditional deep learning methods usually require large amounts of annotated data\, posing a significant challenge in medical imaging due to complex data collection and high annotation costs. Furthermore\, privacy and security concerns restrict data sharing and collaboration between institutions.Federated learning (FL) and annotation-efficient techniques have emerged to address these issues. This seminar explores how combining federated learning with annotation-efficient methods can advance intelligent medical image analysis. The key topics include: 1. Annotation-efficient strategies: Discussing self-supervised and weakly-supervised learning methods to enhance training efficiency and model performance when labeled data is limited; 2. Federated learning applications: Exploring how federated learning enables distributed model training across multiple institutions without data sharing\, thereby protecting data privacy; 3. Practical applications and challenges: Analyzing specific scenarios such as disease diagnosis and organ segmentation\, discussing the strengths and limitations of federated learning and annotation-efficient techniques\, and forecasting future developments. \nSpeaker\nMr. LIN Li\nDepartment of Electrical and Electronic Engineering\nThe University of Hong Kong \nSpeaker’s Biography\nLi LIN received his B.S. in Telecommunication Engineering from South China Normal University in 2018 and the M.S. in Information and Communication Engineering from Sun Yat-sen University in 2021. He is currently pursuing the Ph.D. degree in the Department of Electrical and Electronic Engineering at the University of Hong Kong\, Hong Kong. \nAll are welcome!
URL:https://ece.hku.hk/events/20250512-1/
LOCATION:Online via Zoom
CATEGORIES:Highlights,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:20250513T143000
DTEND;TZID=Asia/Hong_Kong:20250513T153000
DTSTAMP:20260509T183754
CREATED:20250603T041854Z
LAST-MODIFIED:20250603T041854Z
UID:111581-1747146600-1747150200@ece.hku.hk
SUMMARY:Mixture of Experts-augmented Deep Unfolding for Activity Detection
DESCRIPTION:Zoom Link: https://hku.zoom.us/j/95300634244 \nAbstract\nIn the realm of activity detection for massive machine-type communications\, intelligent reflecting surfaces (IRS) have shown significant potential in enhancing coverage for devices lacking direct connections to the base station (BS). However\, traditional activity detection methods are typically designed for a single type of channel model\, which does not reflect the complexities of real-world scenarios\, particularly in systems incorporating IRS. To address this challenge\, this paper introduces a novel approach that combines model-driven deep unfolding with a mixture of experts (MoE) framework. By automatically selecting one of three expert designs and applying it to the unfolded projected gradient method\, our approach eliminates the need for prior knowledge of channel types between devices and the BS. Simulation results demonstrate that the proposed MoE-augmented deep unfolding method surpasses the traditional covariance-based method and black-box neural network design\, delivering superior detection performance under mixed channel fading conditions. \nSpeaker\nMr. REN Zeyi\nDepartment of Electrical and Electronic Engineering\nThe University of Hong Kong \nSpeaker’s Biography\nZeyi Ren received the B.Eng. degree from Beijing Institute of Technology\, Beijing\, China\, in 2023. He is currently working toward the M.Phil. degree with The University of Hong Kong\, Hong Kong. His research interests include model driven deep learning and wireless communications. \nAll are welcome!
URL:https://ece.hku.hk/events/20250513-1/
LOCATION:Online via Zoom
CATEGORIES:Highlights,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:20250515T150000
DTEND;TZID=Asia/Hong_Kong:20250515T160000
DTSTAMP:20260509T183754
CREATED:20250603T034632Z
LAST-MODIFIED:20250603T034632Z
UID:111567-1747321200-1747324800@ece.hku.hk
SUMMARY:Distributed Mixture-of-Expert Systems at the Wireless Edge (Duplicate)
DESCRIPTION:Zoom Link: https://hku.zoom.us/j/91757354553?pwd=tHpInMTglaIVMJLek0ydP0vddHihh8.1 \nMeeting ID: 917 5735 4553\nPasscode: 587193 \nAbstract\nExisting Video-to-Audio (V2A) models typically generate sound based solely on visual input\, offering limited user control. To address this limitation\, we propose a multimodal controllable V2A system that conditions audio generation on a variety of user inputs– such as text\, images\, or audio– in addition to video. Our approach leverages the ImageBind model to align these diverse input modalities into a shared representation\, which is then used to guide audio generation. \nDue to the lack of multimodal datasets for audio generation\, we constructed a training dataset comprising large-scale text-audio data complemented with a limited amount of video-audio data to enable controllable and context-aware audio generation. To further enhance generative quality\, we introduce several data ensemble strategies: (1) Source Balancing\, which maintains a trade-off between concept diversity and sample diversity\, and (2) two synchronization techniques– Audio Feature Selector (AFS) and Audio Peak IoU Matching (APIM)– to improve temporal alignment between video and generated audio. \nOur system enables flexible and precise audio generation that aligns closely with multimodal user intent. Finally\, we introduce a novel benchmark with a cross-sample evaluation framework\, designed to standardize assessments of multimodal V2A systems by evaluating consistency and diversity across input-output combinations. Our method achieves state-of-the-art performance\, demonstrating significant improvements in audio quality\, synchronization\, and input-condition alignment. \nSpeaker\nMr. HE Ruifei\nDepartment of Electrical and Electronic Engineering\nThe University of Hong Kong \nSpeaker’s Biography\nMr. Ruifei He is a final-year Ph.D. student in Department of Electrical and Electronic Engineering at the University of Hong Kong. He obtained his B.Eng. degree in the Department of Automation at Zhejiang University. His research focuses on data-centric/efficient learning (e.g. synthetic/generative data\, mixing multi-modal data\, and semi-supervised learning) for computer vision tasks. \n\nAll are welcome!
URL:https://ece.hku.hk/events/20250515-1/
LOCATION:Online via Zoom
CATEGORIES:Highlights,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:20250515T150000
DTEND;TZID=Asia/Hong_Kong:20250515T160000
DTSTAMP:20260509T183754
CREATED:20250603T035131Z
LAST-MODIFIED:20250603T035131Z
UID:111572-1747321200-1747324800@ece.hku.hk
SUMMARY:End-to-end High-quality Posterior Ocular Shape Reconstruction in Ophthalmology
DESCRIPTION:Zoom Link: https://hku.zoom.us/j/98147018160?pwd=KEuU1XQtISq3HQpWJyF6itZMJ1hYY5.1\nMeeting ID: 981 4701 8160\nPassword: 294231 \nAbstract\nAccurately estimating morphological changes of the Posterior Eyeball Shape (PES) is a critical task in ophthalmology\, since the PES is a crucial factor in many clinical applications\, such as myopia prevention\, surgical planning\, and disease screening. However\, existing imaging devices are constrained by limited field-of-view (FOV) and insufficient resolution\, thus providing insufficient diagnostic information for surgeons to make accurate decisions. Previous segment-based reconstruction methods suffer from two main drawbacks: first\, common imaging modalities can’t provide intact enough shape details as constraints\, thus requiring intricate pre- and post-processing; second\, existing data representations struggle to trade-off between computational efficiency and reconstruction quality\, thus hindering end-to-end reconstruction with fine-grained shape details. \nBenefiting from our more efficient 2D representation in polar coordinate\, we propose a novel task of reconstructing intact 3D PES based on purely small-FOV OCT scans and introduces a novel Posterior Eyeball Shape Network (PESNet) to accomplish this task. The proposed PESNet equips the Siamese structure that incorporates anatomical information of the eyeball as guidance. To capture more detailed information\, we introduce a Polar Voxelization Block (PVB) that transfers sparse input point clouds to a dense representation. Furthermore\, we propose a Radius-wise Fusion Block (RFB) that fuses correlative hierarchical features from the two branches. Finally\, this high-cost reconstruction task is compressed into the 2D surface map regression task. The experiments indicate that our method achieves state-of the-art performance\, providing a well-represented complete posterior eyeball shape on both healthy and patient cases. This result demonstrates that our method offers a significant improvement over existing methods in accurately reconstructing the complete 3D posterior eyeball shape. This achievement has important implications for clinical applications. \nSpeaker\nMr. ZHANG Jiaqi\nDepartment of Electrical and Electronic Engineering\nThe University of Hong Kong \nSpeaker’s Biography\nMr. Jiaqi Zhang obtained his B.Sc. degree from Northeastern University (NEU) in China and his M.Res. degree from the National University of Singapore (NUS) in Singapore. He is currently pursuing the Ph.D. in the Department of Electrical and Electronic Engineering at the University of Hong Kong (HKU)\, under the supervision of Prof. Xiaojuan Qi. His research focuses on medical image processing\, 3D reconstruction\, AIGC\, and representation learning. \n\nAll are welcome!
URL:https://ece.hku.hk/events/20250515-2/
LOCATION:Online via Zoom
CATEGORIES:Highlights,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:20250519T110000
DTEND;TZID=Asia/Hong_Kong:20250519T120000
DTSTAMP:20260509T183754
CREATED:20250603T024808Z
LAST-MODIFIED:20250603T025046Z
UID:111483-1747652400-1747656000@ece.hku.hk
SUMMARY:Symmetric Diffusers: Learning Discrete Diffusion on Finite Symmetric Groups
DESCRIPTION:We regret to inform you that the event has been cancelled and will be postponed to a later date.  \nAbstract\nFinite symmetric groups Sn are essential in fields such as combinatorics\, physics\, and chemistry. However\, learning a probability distribution over Sn poses significant challenges due to its intractable size and discrete nature. We introduce SymmetricDiffusers\, a novel discrete diffusion model that simplifies the task of learning a complicated distribution over Sn by decomposing it into learning simpler transitions of the reverse diffusion using deep neural networks. We identify the riffle shuffle as an effective forward transition and provide empirical guidelines for selecting the diffusion length based on the theory of random walks on finite groups. Additionally\, we propose a generalized Plackett-Luce (PL) distribution for the reverse transition\, which is provably more expressive than the PL distribution. We further introduce a theoretically grounded “denoising schedule” to improve sampling and learning efficiency. Extensive experiments show that our model achieves state-of-the-art or comparable performances on solving tasks including sorting 4-digit MNIST images\, jigsaw puzzles\, and traveling salesman problems. \nSpeaker\nProf. Renjie LIAO\nDepartment of Electrical and Computer Engineering\, and\nDepartment of Computer Science\,\nUniversity of British Columbia (UBC) \nSpeaker’s Biography\nRenjie Liao is an Assistant Professor in the Department of Electrical and Computer Engineering and an Associate Member of the Department of Computer Science at the University of British Columbia (UBC). He is also a faculty member at the Vector Institute and holds a Canada CIFAR AI Chair. Prior to joining UBC\, he was a Visiting Faculty Researcher at Google Brain\, working with Geoffrey Hinton and David Fleet. He received his Ph.D. in Computer Science from the University of Toronto in 2021\, under the supervision of Richard Zemel and Raquel Urtasun. During his Ph.D.\, he also worked as a Senior Research Scientist at Uber Advanced Technologies Group. He holds an M.Phil. in Computer Science from the Chinese University of Hong Kong (2015) and a B.Eng. in Automation from Beihang University (2011). His research interests span machine learning and its intersection with computer vision\, self-driving\, healthcare\, and beyond\, with a particular focus on probabilistic and geometric deep learning. \nOrganiser\nProf. Xiaojuan QI\nDepartment of Electrical and Electronic Engineering\,\nThe University of Hong Kong \nAll are welcome!
URL:https://ece.hku.hk/events/20250519-1/
LOCATION:Room CB-603\, 6/F\, Chow Yei Ching Building\, The University of Hong Kong
CATEGORIES:Highlights,Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20250519T153000
DTEND;TZID=Asia/Hong_Kong:20250519T163000
DTSTAMP:20260509T183754
CREATED:20250603T034323Z
LAST-MODIFIED:20250603T034323Z
UID:111564-1747668600-1747672200@ece.hku.hk
SUMMARY:Distributed Mixture-of-Expert Systems at the Wireless Edge
DESCRIPTION:Zoom Link: https://hku.zoom.us/j/93486553339 \nAbstract\nThe emergence of distributed Mixture-of-Experts (DMoE) systems\, which deploy expert models at edge nodes\, offers a pathway to achieving connected intelligence in sixth-generation (6G) mobile networks and edge artificial intelligence (AI). However\, current DMoE systems lack an effective expert selection algorithm to address the simultaneous task-expert relevance and channel diversity inherent in these systems. Traditional AI or communication systems focus on either performance or channel conditions\, and direct application of these methods leads to high communication overhead or low performance. To address this\, we propose the DMoE protocol to schedule the expert inference and inter-expert transmission. This protocol identifies expert selection and subcarrier allocation as key optimization problems. We formulate an expert selection problem by incorporating both AI performance and channel conditions\, and further extend it to a Joint Expert and Subcarrier Allocation (JESA) problem for comprehensive AI and channel management within the DMoE framework. For the NP-hard expert selection problem\, we introduce the Dynamic Expert Selection (DES) algorithm\, which leverages a linear relaxation as a bounding criterion to significantly reduce search complexity. For the JESA problem\, we discover a unique structural property that ensures asymptotic optimality in most scenarios. We propose an iterative algorithm that addresses subcarrier allocation as a subproblem and integrates it with the DES algorithm. The proposed framework effectively manages the tradeoff between task relevance and channel conditions through a tunable importance factor\, enabling flexible adaptation to diverse scenarios. Numerical experiments validate the dual benefits of the proposed expert selection algorithm: high performance and significantly reduced cost. JESA consistently achieves higher accuracy compared to homogeneous expert selection and lowers the cost by up to 50% compared to Top-k scheduling. \nSpeaker\nMr. Shengling Qin\nDepartment of Electrical and Electronic Engineering\nThe University of Hong Kong \nSpeaker’s Biography\nShengling Qin received the B.Eng. degree from Tsinghua University\, China\, in 2023. He is currently working towards MPhil degree with The University of Hong Kong\, Hong Kong. His research interests include mixture-of-experts\, large language models and distributed training. \n\n\n\nAll are welcome!
URL:https://ece.hku.hk/events/20250519-2/
LOCATION:Online via Zoom
CATEGORIES:Highlights,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:20250519T160000
DTEND;TZID=Asia/Hong_Kong:20250519T170000
DTSTAMP:20260509T183754
CREATED:20250603T024619Z
LAST-MODIFIED:20250603T041708Z
UID:111476-1747670400-1747674000@ece.hku.hk
SUMMARY:Strategies on Perovskite Nanocrystals for Achieving High-Performance Light Emitting Devices
DESCRIPTION:Zoom Link: https://hku.zoom.us/j/92719085356?pwd=09aQ3vjvg9bhXcObBYxjNtj4UVx5V4.1 \n\nAbstract\nMixed-chloride/bromide perovskite nanocrystals (PeNCs) are known for their advantages in pure blue emission\, but often suffer from halogen segregation. This study investigates the ligand exchange process with different ion pair combinations to improve stability. Surprisingly\, altering the ligand ion combinations leads to a deviation from pure blue emission in CsPbBrxCl3-x nanocrystals due to halogen redistribution influenced by solubility principles in a non-polar environment. Furthermore\, a novel approach is demonstrated in this study by capping p-type cuprous sulfide (Cu2S) over Cs3Cu2I5 to enhance hole mobility in Cu-based perovskite nanocrystals. The resulting Cs3Cu2I5/Cu2S nanocrystals exhibit improved hole mobility and photoluminescence quantum yield\, leading to enhanced electroluminescent performance in white perovskite light-emitting diodes (W-PeLEDs). \nSpeaker\nMr. LI Dongyu\nDepartment of Electrical and Electronic Engineering\nThe University of Hong Kong \nSpeaker’s Biography\nLI Dongyu received the M.S. degree from Jilin University in 2019. He is currently pursuing his Ph.D. degree with the Department of Electrical and Electronic Engineering\, The University of Hong Kong. His research interest is semiconductor materials and their application in light-emitting devices. \nAll are welcome!
URL:https://ece.hku.hk/events/20250519-01/
LOCATION:Online via Zoom
CATEGORIES:Highlights,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:20250520T100000
DTEND;TZID=Asia/Hong_Kong:20250520T110000
DTSTAMP:20260509T183754
CREATED:20250603T024946Z
LAST-MODIFIED:20250603T025050Z
UID:111488-1747735200-1747738800@ece.hku.hk
SUMMARY:Keynode-Driven Dynamic Mesh Compression
DESCRIPTION:Abstract\n3D Dynamic Meshes can deliver engaging experiences in various applications\, but the storage and transmission demands associated with these data structures can be prohibitive. We address this challenge with an efficient compression technique leveraging embedded key nodes. The temporal motion of each vertex is formulated as a distance-weighted combination of transformations from neighboring key nodes\, requiring the transmission of solely the key nodes’ transformations. Through extensive experiments\, we demonstrate the effectiveness of our method in significantly reducing storage requirements while preserving the immersive quality of the visual content. \nSpeaker\nProf. Truong NGUYEN\nElectrical and Computer Engineering\,\nUniversity of California San Diego \nBiography of the Speaker\nTruong Nguyen received his B.S.\, M.S. and Ph.D. in Electrical Engineering at California Institute of Technology in 1985\, 1986 and 1989. His current research areas include 3D Human Mesh segmentation and coding as well as machine learning for medical image analysis. During his academic career\, he has published over 200 peer-reviewed journal papers\, over 380 peer-reviewed conference papers\, 1 textbook\, 3 book chapters\, and 15 issued patents. His Google H-index is 72 with over 31K citations. He received the NSF Career Award in 1995\, and IEEE Signal Processing Society’s 1992 Paper Award (student author) for the 1990 paper “Structures for M-Channel Perfect-Reconstruction FIR QMF Banks Which Yield Linear-Phase Analysis Filters” (IEEE Trans. ASSP). He received the Distinguished Teaching Award at UCSD in 2019 and three teaching awards from the ECE Dept. at UCSD in 2006\, 2008 and 2010. He is a Fellow of IEEE. \nOrganiser\nProf. S.C. CHAN\nDepartment of Electrical and Electronic Engineering\,\nThe University of Hong Kong \nCo-organiser\nIEEE Signal Processing Society Hong Kong Chapter \nAll are welcome!
URL:https://ece.hku.hk/events/20250520-1/
LOCATION:Room CB-603\, 6/F\, Chow Yei Ching Building\, The University of Hong Kong
CATEGORIES:Highlights,Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2025/06/1280-3.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20250520T140000
DTEND;TZID=Asia/Hong_Kong:20250520T150000
DTSTAMP:20260509T183754
CREATED:20250603T032657Z
LAST-MODIFIED:20250603T032657Z
UID:111558-1747749600-1747753200@ece.hku.hk
SUMMARY:Understanding Complex-Valued Transformer for Modulation Recognition
DESCRIPTION:Zoom Link: https://hku.zoom.us/j/95380440070 \nAbstract\nComplex-valued convolution neural networks (CVCNNs) have been recently applied for modulation recognition (MR)\, due to its ability to capture the relationship between the real and imaginary parts of the received signal. On the other hand\, the transformer model has been shown to be distinguished in MR by its superior capability to extract the correlation among high-dimensional signals compared to the CNN. It is a logical next step to ask whether a fully complex-valued transformer based neural network (CVTNN) can bring further performance gain? If so\, where the gain comes from? To answer these questions\, this letter designs the building blocks of the CVTNN for MR\, which is composed of a convolution embedding module\, a complete transformer encoder\, and a C2R classifier\, and establishes the estimation error bound of the proposed CVTNN from an inductive bias perspective. We theoretically prove that the estimation error bound of the proposed CVTNN is lower than that of the real-valued transformer based neural network (RVTNN) for MR. Simulation results further show that the proposed CVTNN outperforms the RVTNN and other benchmarks under different settings\, which corroborates the proposed theoretical analysis. \nSpeaker\nMr. Jingreng Lei\nDepartment of Electrical and Electronic Engineering\nThe University of Hong Kong \nSpeaker’s Biography\nJingreng Lei received the B.Eng. degree from Sun Yat-sen University\, China\, in 2023. He is currently working towards MPhil degree with The University of Hong Kong\, Hong Kong. His research interests include complex-valued neural network\, distributed optimization and wireless communication. \nAll are welcome!
URL:https://ece.hku.hk/events/20250520-3/
LOCATION:Online via Zoom
CATEGORIES:Highlights,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:20250522T143000
DTEND;TZID=Asia/Hong_Kong:20250522T160000
DTSTAMP:20260509T183754
CREATED:20250603T025446Z
LAST-MODIFIED:20250603T025451Z
UID:111502-1747924200-1747929600@ece.hku.hk
SUMMARY:Reimagining Edge AI and LLM Inference with Compute Memory Architectures
DESCRIPTION:Abstract\nRecent advances in artificial intelligence (AI)\, especially in large language models (LLMs)\, have dramatically increased model sizes and computational demands\, significantly straining computing system capabilities. This issue is particularly acute in resource-constrained edge AI scenarios\, where efficient hardware acceleration of compute-intensive tasks and optimization of data reuse to minimize costly data transfers are essential. Addressing these challenges\, this talk will explore various options for designing compute memory subsystems through innovative circuit-level and system-level approaches to enhance the efficiency of edge AI applications. Firstly\, we will introduce MAXWELL\, a near-SRAM co-design computing architecture specifically tailored for edge AI. MAXWELL optimizes performance and energy efficiency by leveraging the regular structure of memory arrays\, achieving high parallelization for both convolutional and fully connected layers\, while supporting fine-grained quantization for real-time image and video processing\, autonomous vehicles\, and Internet of Things (IoT) devices. Secondly\, we will delve into SLIM\, a complementary approach to MAXWELL for big data analytics\, scientific computing\, and financial modeling that provides a cost-effective solution for sparse LLM inference. SLIM leverages in-memory processing in DRAM to significantly reduce latency for large caching datasets in edge AI and utilizes near-storage in Solid State Drives (SSD) for large LLM models. Potential applications of SLIM include. This talk will demonstrate that these complementary approaches for compute memory subsystems can achieve up to 10x speed-ups compared to state-of-the-art edge AI accelerators that require data transfers at the boundaries of ML layers. Furthermore\, the proposed co-design approach can accelerate performance by up to 250x compared to pure software optimizations on the X-HEEP edge AI open-source platform\, which integrates MAXWELL and SLIM logic with a 32-bit RISC-V core. Notably\, these accelerator-specific components of computing memories account for less than 12% of the total memory area of X-HEEP. \nSpeaker\nProf. David ATIENZA\nFull Professor of Electrical and Computer Engineering\,\nHead of the Embedded Systems Laboratory (ESL)\,\nAssociate Vice President for Research Centers and Technology Platforms\,\nÉcole polytechnique fédérale de Lausanne (EPFL)\, Switzerland \nSpeaker’s Biography\nDavid Atienza is a Full Professor of Electrical and Computer Engineering at EPFL\, Switzerland\, where he heads the Embedded Systems Laboratory (ESL) and serves as the Associate Vice President for Research Centers and Technology Platforms. His research focuses on system-level design methodologies for energy-efficient multi-processor system-on-chip (MPSoC) architectures\, targeting next-generation computing servers\, data centers\, and edge AI embedded systems\, particularly smart wearables and medical devices in the Internet of Things (IoT) era. Dr. Atienza has co-authored over 450 publications\, holds 14 patents\, and has received numerous best paper awards at top conferences. He is the Editor-in-Chief of IEEE TCAD and ACM CSUR\, and has served as the Technical Program Chair of DATE 2015 and General Chair of DATE 2017. Among other awards\, he has received the 2024 Test-of-Time Best Paper Award at the IEEE/ACM CODES+ISSS Conference\, the IEEE/ACM ICCAD 10-Year Retrospective Most Influential Paper Award in 2020\, the DAC Under-40 Innovators Award in 2018\, and an ERC Consolidator Grant in 2016. He has also served as President of IEEE CEDA (2018-2019) and Chair of the EDAA (2022-2024). He is a Fellow of IEEE and ACM. \nOrganiser\nProf. Kaibin HUANG\nDepartment of Electrical and Electronic Engineering\,\nThe University of Hong Kong \nAll are welcome!
URL:https://ece.hku.hk/events/20250522-1/
LOCATION:Room CB-603\, 6/F\, Chow Yei Ching Building\, The University of Hong Kong
CATEGORIES:Highlights,Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2025/06/1280-4.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20250523T160000
DTEND;TZID=Asia/Hong_Kong:20250523T160000
DTSTAMP:20260509T183754
CREATED:20250603T032424Z
LAST-MODIFIED:20250603T032435Z
UID:111553-1748016000-1748016000@ece.hku.hk
SUMMARY:Security and Efficient Brain-inspired In-memory Computing
DESCRIPTION:Zoom Link: https://hku.zoom.us/j/97430126742?pwd=ou6CUPNMjhlrmRbwUKRa8aTHi6PjYX.1\nMeeting ID: 974 3012 6742\nPassword: 967270 \nAbstract\nThe human brain operates as a sophisticated spiking neural network (SNN)\, capable of learning multimodal signals in a zero-shot manner by leveraging prior knowledge. Impressively\, it accomplishes this with minimal energy consumption\, relying on event-driven signals that travel through its intricate structure. However\, replicating the brain’s functionality in efficient neuromorphic hardware poses significant challenges in both hardware and software. Moreover\, training these algorithms demands extensive resources\, and effective security measures remain insufficient. \nBenefiting from the RRAM array inherit stochasticity\, we demonstrated an efficient analogue-digital system that can handle multi-modal spiking signals and possess zero-shot learning capability like a human. This reservoir accelerated system enables significant lower training overheads while maintaining comparable baseline utility. Since emerging brain-inspired computing raises security concerns\, we also share new methodologies insights onto these neuromorphic systems that can secure non-volatile CIM-based parameters without sacrificing latency and energy efficiency. \nThis presentation will delve into the development of a secure and efficient brain-inspired in-memory computing system\, achieved through the integrated co-design of algorithms\, circuits\, and devices. \nSpeaker\nMr. WONG Edwin Kwun Hang\nDepartment of Electrical and Electronic Engineering\nThe University of Hong Kong \nSpeaker’s Biography\nMr. Edwin Kwun Hang Wong received the B.Eng. (EE) degree from The University of Hong Kong in 2023. He is currently working towards MPhil degree with The University of Hong Kong. His research interests include AI Security\, Brain-inspired computing\, and RRAM-based accelerator. \nAll are welcome!
URL:https://ece.hku.hk/events/20250523-1/
LOCATION:Online via Zoom
CATEGORIES:Highlights,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:20250527T150000
DTEND;TZID=Asia/Hong_Kong:20250527T160000
DTSTAMP:20260509T183754
CREATED:20250603T032119Z
LAST-MODIFIED:20250603T032140Z
UID:111548-1748358000-1748361600@ece.hku.hk
SUMMARY:Digital Over-the-Air Computation: Achieving High Reliability via Bit-Slicing
DESCRIPTION:Zoom Link: https://hku.zoom.us/j/99273671426?pwd=bm3VeyFWXnLAlUIBBXJDGAmMfzoKJ5.1 \nAbstract\n6G mobile networks aim to realize ubiquitous intelligence at the network edge via distributed learning\, sensing\, and data analytics. Their common operation is to aggregate high-dimensional data\, which causes a communication bottleneck that cannot be resolved using traditional orthogonal multi-access schemes. A promising solution\, called over-the-air computation (AirComp)\, exploits channels’ waveform superposition property to enable simultaneous access\, thereby overcoming the bottleneck. Nevertheless\, its reliance on uncoded linear analog modulation exposes data to perturbation by noise and interference. Hence\, the traditional analog AirComp falls short of meeting the high-reliability requirement for 6G. Overcoming the limitation of analog AirComp motivates this work\, which focuses on developing a framework for digital AirComp. The proposed framework features digital modulation of each data value\, integrated with the bit-slicing technique to allocate its bits to multiple symbols\, thereby increasing the AirComp reliability. To optimally detect the aggregated digital symbols\, we derive the optimal maximum a posteriori detector that is shown to outperform the traditional maximum likelihood detector. Furthermore\, a comparative performance analysis of digital AirComp with respect to its analog counterpart with repetition coding is conducted to quantify the practical signal-to-noise ratio (SNR) regime favoring the proposed scheme. On the other hand\, digital AirComp is enhanced by further development to feature awareness of heterogeneous bit importance levels and its exploitation in channel adaptation. Lastly\, simulation results demonstrate the achivability of substantial reliability improvement of digital AirComp over its analog counterpart given the same channel uses. \nSpeaker\nMr. Jiawei LIU\nDepartment of Electrical and Electronic Engineering\nThe University of Hong Kong \nSpeaker’s Biography\nJiawei Liu received the B.Eng. degree from the Southern University of Science and Technology\, Shenzhen\, China\, in 2020. He is currently pursuing the Ph.D. degree with the Department of Electrical and Electronics Engineering\, The University of Hong Kong (HKU)\, Hong Kong. His research interests include wireless communication system design\, in-memory computing hardware\, and edge intelligence systems. \nAll are welcome!
URL:https://ece.hku.hk/events/20250527-1/
LOCATION:Online via Zoom
CATEGORIES:Highlights,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:20250528T143000
DTEND;TZID=Asia/Hong_Kong:20250528T153000
DTSTAMP:20260509T183754
CREATED:20250603T025615Z
LAST-MODIFIED:20250626T083737Z
UID:111508-1748442600-1748446200@ece.hku.hk
SUMMARY:Seminar on Human-AI Ecosystems for Daily Health and Well-being
DESCRIPTION:Abstract\nAs the intelligence of everyday smart devices continues to evolve\, they can already monitor basic health behaviors such as physical activities and heart rates. The vision of an intelligent health monitoring and intervention pipeline seems to be within reach. How do we get there? \nIn this talk\, I will introduce a comprehensive pipeline that connects AI\, end-users\, and health experts. For end-users\, I will introduce our work that bridges behavior science theory-driven intervention designs and generalizable behavior models. I will also introduce my efforts on passive sensing datasets\, human-centered algorithms & large language models (LLMs)\, as well as a benchmark platform that drives the community toward more robust and deployable health systems for both end-users and experts. \nSpeaker\nProf. Xuhai (Orson) XU\nAssistant Professor\,\nColumbia University \nSpeaker’s Biography\nXuhai (Orson) XU is an Assistant Professor at Columbia University\, Department of Biomedical Informatics and Department of Computer Science (by courtesy)\, where he directs the SEA (Sense\, Empower\, and Augment) Lab. He is also a visiting faculty researcher at Google. He received his PhD at the University of Washington in 2023 and was a postdoc at MIT until 2024. Specializing in human-computer interaction\, applied machine learning\, and health\, Xu develops deployable behavior modeling algorithms to monitor various health and well-being conditions using everyday sensor data and health records. He further designs and deploys intelligent intervention & interaction techniques that help users achieve personal health and well-being goals and support health experts in making decisions. Xu has earned several awards\, including several Best Paper\, Best Paper Honorable Mention\, and Best Artifact awards. His research has been covered by media outlets such as the Washington Post and ACM News. He was recognized as the Outstanding Student Award Winner at UbiComp 2022\, the 2023 UW Distinguished Dissertation Award\, and the 2024 Innovation and Technology Award at the Western Association of Graduate Schools. \nSEA Lab is actively hiring PhD students with the background of HCI\, mobile sensing\, and applied AI\, with the focus in health applications. \nOrganiser\nProf. Edith C.H. NGAI\nDepartment of Electrical and Electronic Engineering\,\nThe University of Hong Kong \nAll are welcome!
URL:https://ece.hku.hk/events/20250528-2/
LOCATION:Room CB-603\, 6/F\, Chow Yei Ching Building\, The University of Hong Kong
CATEGORIES:Highlights,Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2025/06/1280-5.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20250529T103000
DTEND;TZID=Asia/Hong_Kong:20250529T113000
DTSTAMP:20260509T183754
CREATED:20250603T025752Z
LAST-MODIFIED:20250603T025757Z
UID:111512-1748514600-1748518200@ece.hku.hk
SUMMARY:Seminar on Terahertz Optoelectronics for Non-Invasive Imaging and Beyond
DESCRIPTION:Abstract\nTerahertz (THz) imaging technology is growing rapidly due to its potential applications in material exploration\, non-destructive evaluation\, industrial inspection\, and bioinformatics. However\, the practical feasibility of THz imaging systems is significantly constrained by the low efficiency of active THz devices\, long imaging acquisition time\, insufficient use of THz signal datasets\, and their bulky nature. In this talk\, I will present our recent research on high-precision THz imaging systems\, starting from material development\, THz optoelectronics designs\, and system integration toward image reconstruction modalities for on-site applications. As the image data quality and data acquisition speed highly rely on the brightness of THz sources\, we have developed high-performance THz plasmonic photoconductive sources generating mW-level radiating power over a several-THz spectral range\, which offers excellent time-resolved raw data for further image restoration and reconstruction. I will further introduce some of our image reconstruction approaches – equalized compressed sensing imaging\, multi-scale deep-learning fusion imaging\, and compressive hybrid neural network – that further speed up the data acquisition process and achieve significantly better reconstructed imaging quality compared with conventional THz CT modalities. This paves the way toward real-time\, hyperspectral THz 3D imagers in the near future\, which opens the door for various exciting applications in non-destructive sensing\, imaging\, and material inspection. \nSpeaker\nProf. Shang-Hua (Steve) YANG\nAssociate Professor\,\nDepartment of Electrical Engineering\,\nNational Tsing Hua University \nSpeaker’s Biography\nShang-Hua (Steve) YANG is an Associate Professor in the Department of Electrical Engineering at National Tsing Hua University. He is renowned for his significant contributions to THz optoelectronics\, communication\, imaging\, and innovative plasmonic photonics applications. His research findings are published in over 100 refereed papers in peer-reviewed journals and conference proceedings. He has received several prestigious awards\, including the IEEE Antennas and Propagation Society Doctoral Research Award\, MOST Young Scholar Fellowship\, NTHU Young Faculty Research Award\, Human Frontier Science Program Research Grant Award\, and Ta-You Wu Memorial Award (2024). He currently serves as the Director of the NTHU THz Optics & Photonics Center\, Taiwan’s first dedicated THz research center. He is a senior member of IEEE\, Optica\, and SPIE. \nOrganiser\nProf. Kenneth K.Y. WONG\nDepartment of Electrical and Electronic Engineering\,\nThe University of Hong Kong \nAll are welcome!
URL:https://ece.hku.hk/events/20250529-1/
LOCATION:Room CB-603\, 6/F\, Chow Yei Ching Building\, The University of Hong Kong
CATEGORIES:Highlights,Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2025/06/1280-6.jpg
END:VEVENT
END:VCALENDAR