Zhidi LIN

Hi there!

I’m Zhidi Lin (林志地), an Assistant Professor in the Department of Mathematics and Information Technology (MIT) at The Education University of Hong Kong (EdUHK). Previously, I held postdoctoral research positions at The University of Hong Kong, working with Prof. Edwin Fong, and at the National University of Singapore, working with Prof. Alexandre Thiéry and Prof. Jeremy Heng. My research lies at the intersection of statistical signal processing, probabilistic machine learning, and dynamical systems, with an emphasis on uncertainty-aware modeling, inference, and decision-making.

I received my Ph.D. from The Chinese University of Hong Kong, Shenzhen in 2024, where I was fortunate to be advised by Prof. Feng Yin and Prof. Shuguang (Robert) Cui.

Research Interests:

My work focuses on data-driven modeling and uncertainty quantification, with a focus on:

Dynamical systems (Bayesian inference, data assimilation)
Generative models (Diffusion models, Gaussian processes, VAEs, etc)
High-dimensional statistical learning
Applications: System identification, tracking, fault detection, time-series forecasting, etc

I am passionate about interdisciplinary research and welcome collaborations with statisticians, AI researchers, and signal processing experts. If you’d like to discuss ideas or potential projects, feel free to reach out at linzhidi017@gmail.com!

News:

2026.10: Serving as an Area Chair for ICASSP 2027.
2026.10.02: Joined The Education University of Hong Kong (EdUHK) as an Assistant Professor.
2026.09: Serving as an Area Chair for ICLR 2027.
2026.05: Goal Reviewer @ ICML 2026.
2026.05: SegPVSG: Panoptic Video Scene Graph Generation via Temporal Focusing and Generative Augmentation has been accepted by ICML 2026.
2025.12: Efficient Transformed Gaussian Process State-Space Models for Non-Stationary High-Dimensional Dynamical Systems has been accepted by IEEE Transactions on Signal Processing.
2025.09: "Multi-View Oriented GPLVM: Expressiveness and Efficiency" has been accepted by NeurIPS 2025.
2025.09: Serve as Reviewer for ICLR 2026, AISTATS 2026, ICASSP 2026.
2025.08: Special session "Bridging Signal Processing and Machine Learning with Gaussian Processes," has been accepted for ICASSP 2026. Huge thanks to Prof. Petar M. Djurić and Prof. Feng Yin for co-organizing this session.
2025.07: "Scalable Random Feature Latent Variable Models" has been published in IEEE Transactions on Pattern Analysis and Machine Intelligence.
2025.07: Relocating to HKU
2025.06: Contributed talk/poster @ Bayes Comp 2025,
Ensemble filtering in nonlinear dynamical systems
2025.06: Contributed talk @ Bayesian Methods for Distributional and Semiparametric Regression, Bayes Comp 2025
Towards Flexibility and Learning Efficiency of Gaussian Process State-Space Models
2025.04: "Hybrid Data-Driven SSM for Interpretable and Label-Free mmWave Channel Prediction" has been accepted by IEEE Transactions on Mobile Computing.
2025.03: Gave a talk @ Huawei (Shanghai)
Towards Flexibility and Learning Efficiency of Gaussian Process State-Space Models
2025.01: "Sparsity-Aware Distributed Learning for Gaussian Processes with Linear Multiple Kernel" has been accepted by IEEE Transactions on Neural Networks and Learning Systems.
2024.09: Invited to serve as Reviewer for ICLR 2025, AISTATS 2025, ICASSP 2025.
2024.08: "Ensemble Kalman Filtering Meets Gaussian Process SSM for Non-Mean-Field and Online Inference" has been accepted by IEEE Transactions on Signal Processing.
2024.06: I joined NUS as a Research Fellow
2024.06.13: I successfully defended my PhD
2024.05: One paper was accepted by ICML 2024, and one paper was accepted by IEEE FUSION 2024
2023.02: Paper entitled "Output-Dependent Gaussian Process State-Space Model" has been accepted by IEEE ICASSP 2023.
2022.05: Paper entitled "Gaussian Process Regression with Grid Spectral Mixture Kernel: Distributed Learning for Multidimensional Data" has been accepted by IEEE FUSION 2022.
2022.01.05: I pass the Ph.D. Qualifying Examination and become a Ph.D. candidate.
2021.02: Paper entitled "Graph Neural Network for Large-Scale Network Localization" has been accepted by IEEE ICASSP 2021.
2020.11.06: Review paper (29-page) entitled "FedLoc: Federated Learning Framework for Data-Driven Cooperative Localization and Location Data Processing" has been published in IEEE Open Journal of Signal Processing.
2020.05: Paper entitled "An Interpretable and Sample Efficient Deep Kernel for Gaussian Process" has been accepted by UAI 2020.

Miscellanies:

Languages – Mandarin (native), English (fluent), Hokkien (native)
Sports – Badminton 🏸, Swimming 🏊‍♂️, …
Leadership/Community activities
- Residential Tutor, Diligentia College, CUHK-Shenzhen, China.

Contact:

Email: linzhidi017 [AT] gmail.com
Address: Department of Statistics and Actuarial Science, Run Run Shaw Building, HKU, Hong Kong.