Students and ECRs in AI-PriSec


This is the Students and Early Career Researchers (ECRs) homepage of the AI Privacy and Security Interest Group.

Upcoming Seminars

  • 20 March 2022, 10:00 (HK time)
    Luyi Chang (HLJU)
    DeceFL: A Principled Decentralized Federated Learning Framework
    [WeMeet Registration] [Live Stream]

    Abstract:Federated Learning is a machine learning setting where multiple entities (clients) collaborate in solving a machine learning problem, under the coordination of a central server or service provider. Each client’s raw data is stored locally and not exchanged or transferred; instead, focused updates intended for immediate aggregation are used to achieve the learning objective. However, in this Centralized machine learning framework, a central server is needed for collecting and distributing model information (instead of data itself) from every other client, leading to high communication pressure and high vulnerability when there exists a failure at or attack on the central server. In this talk, the speaker will introduce a principled decentralized federated learning framework that can solve those above problems.

    Bio: Luyi Chang is currently working toward a Master’s Degree in Information Statistics Technology at Heilongjiang University. Her interests mainly include decentralized federated learning and mutual learning.

Past Seminars

  • 13 March 2022, 10:00 (HK time)
    Yuyu Yuan (GDUT)
    Digital Twin for Power Systems

**Abstract:**The power system is the most complex, capital, technology-intensive, and man-made composite system in the world. Correct cognition of power systems is the prerequisite for various power grid operation management and regulation decisions. The nonlinear, high latitude, layered, and distributed characteristics of the power systems are the difficulties faced by power system cognition. The application of digital twins in the power system can provide a reference for power transportation management and regulation decision-making by analyzing a large amount of data and mapping the real-time state of entities. In this talk, we will provide an overview of the DT-based power system.

**Bio:** Yuyu Yuan is currently working toward a B.S. degree at the Guangdong University of Technology. Her interests mainly include Digital Twins, the Internet of Things, Metaverse. In this talk, the speaker will introduce the application and prospect of digital twins in a power system.

  • 6 March 2022, 10:00 (HK time)
    Yukang Shen (SenseTime)
    When Virtual Reality Techniques Meet Metaverse: A Front-end Perspective

**Abstract:**Due to the development of network communication technology, Metaverse has entered the public eye. In this context, Metaverse has become a hot research direction, and related virtual reality technologies AR/VR/MR/XR have also become a research hotspot in the industry. In this talk, I will introduce the associations and differences between AR, VR, MR, and XR from the perspective of common sense in computer graphics, as well as some existing technical problems and corresponding solutions. Finally, a method to quickly build a simple VR APP based on ThreeJS and React will be introduced.

**Bio:** Yukang Shen received the B.Eng. degree in Network Engineering from Heilongjiang University, Harbin, China, in 2020. He is now a front-end engineer at SenseTime.

  • 27 February 2022, 11:00 (HK time)
    Zhe Zhang (HLJU)
    Interpret Federated Learning with Shapley Values

    Abstract: With the continuous breakthroughs in the research and application of federated learning, high-performance, complex algorithms and models generally lack the transparency of decision logic and the interpretability of results. Therefore, it is challenging to deploy federated learning technology in national defense, finance, medical care, law, and cybersecurity that require accurate decision-making. In this talk, the presenter will introduce some methods that Shapley values to explain vertical federated learning.

    Bios: Zhe Zhang received the B.S. degree from the North China University of Water Resources and Electric Power, in 2020. He is currently working toward a Master’s Degree in Information Statistics Technology at Heilongjiang University. His research interests are mainly federated learning, semi-supervised learning, and spiking neural networks.

Organizers