⠀Catherine C. Liu

💥25/6/2025 - 28/6/2025 We warmly congratulate the successful hosting of the IEEE International Conference on Quantum Control, Computing and Learning (qCCL 2025) at The Hong Kong Polytechnic University. As members of the academic community at PolyU, we are excited to witness this emerging platform in quantum technologies take place on our campus.

Our group members, Haoting Wan and Shan Lu, are actively exploring topics in quantum computing and quantum information theory. We look forward to learning from the cutting-edge developments presented at this conference and to leveraging the insights gained to accelerate our research progress in this dynamic field.

恭喜IEEE量子控制、计算与学习国际会议（qCCL 2025）在香港理工大学成功举办，我组成员也正在积极探索量子计算与信息方向。

💥2/3/2025 Congratulations to Miss. Lingxiao Li for the acceptance of the paper, entitled "Huber principal component analysis for large-dimensional factor models", by Journal of Econometrics. (https://www.sciencedirect.com/science/article/abs/pii/S0304407625000478)

This work is based on her MPhil's thesis, supervised by Professor Yong He.

The paper proposes two robust algorithms to estimate factor model while accounting for the heavy-tailed nature of financial data. The first is an iterative Huber regression algorithm (IHR) and the second performs PCA on the weighted sample covariance matrix (HPCA). The shiny point is to replace the squared loss in traditional factor analysis with the Huber loss to achieve robust estimation.

恭喜李凌霄的論文"THuber principal component analysis for large-dimensional factor models"被Journal of Econometrics接收！

💥28/1/2025 Congratulations on the collaboration work with Dr. Lorry Luo from the Department of Systems Engineering, CityU, HKSAR. The paper is entitled "A Graph-Theoretic Approach to Detection of Parkinsonian Freezing of Gait from Videos", accepted by Statistics in Medicine. (https://onlinelibrary.wiley.com/doi/full/10.1002/sim.70020)

The paper proposes a novel graph-theoretic approach for detecting Freezing of Gait (FOG) from video data of patients with Parkinson’s Disease (PD). This approach differs from prevailing pixel-based deep learning methods, offering a novel perspective on feature extraction for FOG detection and potentially contributing to improved diagnosis and treatment of PD. Specifically, the spatial relations and temporal progression of a patient’s posture over time are represented as a sequence of pose graphs, where joint positions serve as nodes, and edges are formed based on anatomical connections and spatial proximity. A hypothesis testing procedure based on Fréchet statistics is proposed to identify temporal breakpoints between regular gait and FOG episodes. The proposed framework is validated on two datasets (Kinect3D and AlphaPose), demonstrating its effectiveness in FOG detection from video data.

Comments from the referees: The proposed method is intriguing, and in my opinion, both effective and efficient. An efficient algorithm based on recursive updates is proposed reducing the computational complexity from O(n^2) to O(n). The authors present a novel graph-theoretic method for detecting freezing of gait (FOG) from video data of Parkinson’s disease (PD) patients. The problem of interest is well-developed and novel, and overall the presentation of the work is interesting.

团队與香港城市大學系統工程系的羅博士合作的论文已被Statistics in Medicine接受發表！

💥11/10/2024 Congratulations to Dr. Xu Zhang for the acceptance of the paper, entitled "Tucker tensor factor models: matricization and mode-wise PCA estimation", by Science China Mathematics. (https://arxiv.org/abs/2206.02508)

Starting from the Tucker tensor factor model, the authors "decode" the estimation of a higher-order tensor factor model by leveraging tensor matricization and summarize that estimations of the low-rank signal components are essentially mode-wise PCA. Based on this mode-wise thought, the projection and iteration-based estimation methods are investigated, and their inferential theory and numerical studies are comprehensively examined.

This work is funded and co-supervised by Guodong Li, HKU.

恭喜張旭博士的論文"Tucker tensor factor models: matricization and mode-wise PCA estimation"被Science China Mathematics接收。

💥26/11/2024 Congratulations to Dr. Chong Zhong for the acceptance of the paper, entitled "CeCNN: Copula-enhanced convolutional neural networks in joint prediction of refraction error and axial length based on ultra-widefield fundus images", by The Annals of Applied Statistics. (https://www.e-publications.org/ims/submission/AOAS/user/submissionFile/66178?confirm=f6b6ca56)

The paper proposes the Copula-enhanced Convolutional Neural Network (CeCNN), a one-stop UWF-based ophthalmic AI framework to jointly predict SE, AL, and myopia status. The shiny point is to model and incorporate the conditional dependence among responses in a CNN through a new copula-likelihood loss. This paper illustrates statistics harnessing AI.

Comments from the referees: “The motivation and the application seem very interesting and relevant. I like the copula idea a lot, which can potentially inspire many other works, even those that are not in this particular application” and “The authors of this paper should be congratulated for working to develop AI systems to improve care for an important clinical problem. ”

This work is supervised by A.H. Welsh, a Fellow of the Australian Academy of Science, the Institute of Mathematical Statistics.

團隊與復旦大學附屬眼耳鼻喉科醫院，復旦大學大數據學院的合作研究被統計期刊The Annals of Applied Statistics接收。鐘充博士與大数据学院的李暘博士平等贡献。

💥17/04/2024, Congratulations to Mr. Qiuyi Huang and Dr. Chong Zhong for the acceptance of their paper, entitled "OUCopula: Bi-Channel Multi-Label Copula-Enhanced Adapter-Based CNN for Myopia Screening Based on OU-UWF Images", in the International Joint Conference on Artificial Intelligence 2024 (IJCAI). (https://www.ijcai.org/proceedings/2024/655)

Comments from the review, “The paper introduces a novel framework called OUCopula for myopia screening, which integrates ultrawideﬁeld imaging with a bi-channel multi-label Convolutional Neural Network (CNN). The proposed framework shows promise in improving myopia screening, particularly through the joint prediction of clinical scores for both eyes. It introduces innovative concepts and methods that could have a significant impact on the AI research community, especially in the ﬁeld of medical imaging.”

團隊與復旦大學附屬眼耳鼻喉科醫院，復旦大學大數據學院的合作研究被國際人工智能聯合會議 2024 (International Joint Conference on Artificial Intelligence, IJCAI)的交叉學科及應用/電腦視覺板塊接收。黃秋逸同學與鐘充博士是共同第一作者。

💥12/04/2024, Congratulations to Dr. Xu Zhang on the acceptance of her paper, entitled "Modeling and Learning on High-Dimensional Matrix-Variate Sequences", in the Journal of the American Statistical Association (https://www.tandfonline.com/doi/full/10.1080/01621459.2024.2344687)

恭喜張旭博士的論文"Modeling and Learning on High-Dimensional Matrix-Variate Sequences"被美國統計學會會刊（JASA）接收。

💥17/03/2024, Congratulations to Shouzheng Chen on receiving his fully-funded PhD offer. He is going to begin his 4-year PhD studies at the APSS department here at PolyU in September 2024. He will focus on research in Environmental Psychology under the supervision of Dr. William Chan.

恭喜陳首錚同學獲得四年制博士獎學金。他將於今年九月在香港理工大學應用社會科學系展開為期4年的博士學習。他會在陳凱榮博士的指導下專注於環境心理學的研究。

💥11/2023, Congratulations to Chong Zhong on gaining his PhD degree this month! He successfully defended his PhD thesis "Nonparametric Bayesian Statistics Harnessing the Forces of Data in Change-point Detection and Survival Analysis" on 18 Sept. 2023. Dr. Zhong has begun his one-year postdoctoral training staying with our team.

恭喜鍾充同學在本月獲得博士學位！鍾充博士于2023年9月18日成功通過博士論文答辯，論文題目為 "Nonparametric Bayesian Statistics Harnessing the Forces of Data in Change-point Detection and Survival Analysis"。钟博士已開始繼續在本团队為期一年的博士后培训。

💥24/08/2023, Dr. Xu Zhang was awarded the Young Scientists Fund by the National Natural Science Foundation of China. Big congratulations to Xu!!

恭喜張旭博士獲得今年的国家自然科学基金青年科学基金项目！！

14/08/2023, SIAT & AMA Workshop on Image Data Analysis.

Google Sites

Report abuse