PeopleofSuperintelligence

Researcher

Ka Ho Chow

Also known as Ka-Ho Chow, 周嘉豪

Ka Ho Chow is a computer scientist at the University of Hong Kong who studies how AI systems can be attacked and defended, with work on federated learning, face recognition and retrieval-augmented generation. He earned a Ph.D. in Computer Science at Georgia Tech under Ling Liu, and the paper LDP-Fed, which he coauthored, won the Best Paper Award at ACM EdgeSys in 2020. A Croucher Scholar in 2021 and an IBM PhD Fellow in 2022, he worked as a research scientist at IBM Research before joining HKU. His papers include STDLens at CVPR 2023, the Chameleon privacy mask at ECCV 2024, Geminio at ICCV 2025 and Protego at CVPR 2026. He is an assistant professor in the School of Computing and Data Science at HKU and an associate editor of IEEE Transactions on Big Data.

Timeline

  1. 2019

    Georgia Tech doctoral studies

    In 2019, he receives the Chair's Fellowship from the School of Computer Science at the Georgia Institute of Technology, where he completes a Ph.D. in Computer Science advised by Ling Liu.

  2. 2020

    Best paper at EdgeSys

    LDP-Fed: Federated Learning with Local Differential Privacy, which he coauthors with Stacey Truex, Ling Liu, Mehmet Emre Gursoy and Wenqi Wei, wins the Best Paper Award at the ACM International Workshop on Edge Systems, Analytics and Networking in 2020.

  3. 2021–2022

    Croucher and IBM fellowships

    In 2021, the Croucher Foundation awards him a Croucher Scholarship for study at Georgia Tech, and in 2022 he is named an IBM PhD Fellow.

  4. 2023

    STDLens against model hijacking

    At CVPR 2023, he and Ling Liu, Wenqi Wei, Fatih Ilhan and Yanzhao Wu publish STDLens, a model hijacking-resilient federated learning method for object detection.

  5. By 2024

    Assistant professor at HKU

    By 2024, he is an assistant professor in the Division of Computer Science at the University of Hong Kong, after working as a research scientist at IBM Research. A Croucher Foundation start-up grant of HKD 500,000 supports his project on trustworthy federated learning against model leakage from 2024 to 2029.

  6. 2024

    The Chameleon privacy mask

    At ECCV 2024, he and Sihao Hu, Tiansheng Huang and Ling Liu present Chameleon, which learns a personalized privacy protection mask (P3-Mask) for each user to block unauthorized face recognition. A related paper, Diversity-driven Privacy Protection Masks Against Unauthorized Face Recognition, appears in PoPETs 2024.

  7. 2025–2026

    Geminio and Protego

    At ICCV 2025, he and Junjie Shan, Ziqi Zhao, Jialin Lu, Rui Zhang and Siu Ming Yiu present Geminio, a gradient inversion attack in federated learning that lets an attacker describe in natural language the data to reconstruct. With Ziling Wang, Shuya Yang and Jialin Lu, he presents Protego, a privacy mask against retrieval-based face recognition, at CVPR 2026 in June 2026, where it receives the CVPR Compute Gold Star. He also coauthors Confundo with Haoyang Hu, Zhejun Jiang and others at USENIX Security 2026, a framework that fine-tunes a large language model to generate poison for retrieval-augmented generation systems.

  8. As of October 2026

    Trustworthy AI at HKU

    He is an assistant professor in the Division of Computer Science, School of Computing and Data Science, at the University of Hong Kong. His research covers security and privacy threats to AI systems and attack-resilient defenses, in areas such as federated learning, large language models and FinTech. In December 2025, he joins the editorial board of IEEE Transactions on Big Data as an associate editor.

Career

  1. NowUniversity of Hong KongAssistant Professor
  2. IBM ResearchResearch Scientist

Education

  1. 2019Georgia Institute of TechnologyPh.D. in Computer Science

From the biography.

  1. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other universityIncludes Ka Ho Chow. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other university. We compare doctorate and 5+ year employment links, career timing and current workplaces for 187 major-prize and 1,111 best-paper winners.
  2. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionIncludes University of Hong Kong. We estimate where the people shaping AI studied and where their careers led. Massachusetts Institute of Technology leads with 467 former students, including 291 recorded doctorates.
  3. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes University of Hong Kong. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.
  4. Google has the highest combined h-index among AI organizationsIncludes University of Hong Kong. We estimate that Google holds the highest combined h-index at 8,016, while universities account for 73% of the deduplicated total across AI organizations.

Sources and credits

Photo: We have no published photo of Ka Ho Chow yet. Send us a free-licensed photo.

  1. Chow, Ka Ho cs.hku.hk. Checked 21 October 2026.
  2. Ka-Ho Chow khchow.com. Checked 21 October 2026.
  3. EdgeSys Programme edge-sys.github.io. Checked 21 October 2026.
  4. Croucher Scholarships | Croucher Foundation scholars.croucher.org.hk. Checked 21 October 2026.
  5. STDLens: Model Hijacking-Resilient Federated Learning for Object Detection openaccess.thecvf.com. Checked 21 October 2026.
  6. AI 'mask' could protect you from facial recognition - Futurity futurity.org. Checked 21 October 2026.
  7. Personalized Privacy Protection Mask Against Unauthorized Facial Recognition arxiv.org. Checked 21 October 2026.
  8. PoPETs Proceedings — Diversity-driven Privacy Protection Masks Against Unauthorized Face Recognition petsymposium.org. Checked 21 October 2026.
  9. ICCV Poster Geminio: Language-Guided Gradient Inversion Attacks in Federated Learning iccv.thecvf.com. Checked 21 October 2026.
  10. CVPR Poster Protego: User-Centric Pose-Invariant Privacy Protection Against Face Recognition-Induced Digital Footprint Exposure cvpr.thecvf.com. Checked 21 October 2026.
  11. Confundo: Learning to Generate Robust Poison for Practical RAG Systems | USENIX usenix.org. Checked 21 October 2026.