About Me
Introduction
My name is Zhong Li, and I am currently a postdoctoral researcher in Computer Science at Leiden University, affiliated with the EDA Lab. I completed my PhD under the supervision of Dr. Matthijs van Leeuwen (daily supervisor and promotor) and Prof. Dr. Thomas Bäck (promotor). In 2024, I was a visiting PhD researcher at the DAML group, Technical University of Munich, supervised by Prof. Dr. Stephan Günnemann.
📘 PhD Thesis: Trustworthy Anomaly Detection for Smart Manufacturing
Research Interests
My research lies broadly in Data Mining and Machine Learning, with a strong focus on Trustworthy AI.
Postdoctoral Research
I am currently working on trustworthy generative models, with a central belief:
When generative models can provide insights into the data they generate, they should also offer insights about themselves.
Key areas of interest include:
- Flow Matching and Diffusion Models: What are they actually learning? How can we better extend these models for tabular and/or discrete data?
- Understanding Unsupervised Learning: Why does self-supervised learning work (or not)? How can it be improved?
PhD Research
During my PhD, I focused on trustworthy anomaly detection, particularly for complex data like event sequences and graph-structured data in smart manufacturing contexts. My goals were to enhance:
- Accuracy: Improve performance with high-dimensional, unstructured, multimodal data
- Explainability: Make models more interpretable and human-friendly
- Generalizability: Ensure robustness under data shifts, unknown hyperparameter settings, and adversarial noise
Recent News
- 2024.11: Presenting our paper Explainable Graph Neural Networks Under Fire at the Amsterdam meetup for the global LoG Conference
- 2024.09: 🔥🔥 TKDE Research paper: Cross-domain Graph Level Anomaly Detection
- 2024.01: 🔥🔥 ICSE’24 Poster: Graph Neural Networks based Log Anomaly Detection and Explanation
- 2023.07: 🔥🔥 TKDD Survey paper: A survey on explainable anomaly detection — 🏆 Highly cited paper based on ESI.
- 2023.07: 🔥🔥 DMKD Research Paper: Explainable Contextual Anomaly Detection using Quantile Regression Forests
- 2022.12: 🔥🔥 SIGKDD Explorations Research paper: Feature Selection for Fault Detection and Prediction based on Event Log Analysis
- 2021.10: 🔥🔥 Clinical Trials Research paper: Choosing and changing the analysis scale in non-inferiority trials with a binary outcome
- [
Position Cancelled] Internship & thesis opportunity with Canon on Inkjet Jet Failure Detection (Oct 2023) - [
Position Filled] Internship & thesis with VDL & ASML on Predictive Maintenance via Log Anomaly Detection (June 2023)
Professional Service
- Reviewer for journals:
- IEEE Transactions on Knowledge and Data Engineering (TKDE)
- Data Mining and Knowledge Discovery (DMKD / DAMI)
- International Journal of Computer Vision (IJCV)
- IEEE Transactions on Emerging Topics in Computational Intelligence (TETCI)
- IEEE Internet of Things Journal (IoT)
- IEEE Intelligent Systems
- AI Communications
- PC Member or Reviewer:
- ECMLPKDD
- KDD
- ICLR
Contact
Office: BE 2.23, Gorlaeus Building, Einsteinweg 55, 2333 CC Leiden, The Netherlands
Email: z.li(at)liacs.leidenuniv.nl