About Me
Introduction
Zhong Li is a postdoc researcher on AI4Math (working with Prof.dr. Zaiwen Wen). He is also a guest researcher in Computer Science at Leiden University, affiliated with the EDA Lab. He completed his PhD under the supervision of Dr. Matthijs van Leeuwen (daily supervisor and promotor) and Prof. Dr. Thomas Bäck (promotor). In 2024, he 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
His research lies broadly in Data Mining and Machine Learning, with strong interest in AI for Math.
Postdoctoral Research
During his postdoc, he works on generative AI for Math, with a particular focus on:
- LLM for Optimization Modeling
- Flow Matching and Diffusion Models
PhD Research
During his PhD, he focused on trustworthy anomaly detection, particularly for complex data like event sequences and graph-structured data in smart manufacturing contexts. His 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
- 2025.11: 🔥🔥 Our research paper Towards Automated Self-Supervised Learning for Truly Unsupervised Graph Anomaly Detection is selected by AAAI journal track for presentation.
- 2025.11: 🔥🔥 One research paper is accepted by AAAI for oral presentation: Learning Subgroups with Maximum Treatment Effects without Causal Heuristics. Congratulations to Lincen.
- 2025.09: 🔥🔥 One research paper is accepted by NeurIPS: Scalable, Explainable and Provably Robust Anomaly Detection with One-Step Flow Matching
- 2025.06: 🔥🔥 One research paper is accepted by DMKD: Towards Automated Self-Supervised Learning for Truly Unsupervised Graph Anomaly Detection
- 2025.05: 🎓 Successfully defended my PhD thesis on Trustworthy Anomaly Detection for Smart Manufacturing at Leiden University!
📄 Read the full thesis - 2024.11: Presenting our paper Explainable Graph Neural Networks Under Fire at the Amsterdam meetup for the global LoG Conference
- 2024.09: 🔥🔥 One research paper is accepted by TKDE: Cross-domain Graph Level Anomaly Detection
- 2024.01: 🔥🔥 ICSE’24 Poster: Graph Neural Networks based Log Anomaly Detection and Explanation
- 2023.07: 🔥🔥 One survey paper is accepted by TKDD: A survey on explainable anomaly detection — 🏆 Highly cited paper based on ESI.
- 2023.07: 🔥🔥 One research paper is accepted by DMKD: Explainable Contextual Anomaly Detection using Quantile Regression Forests
- 2022.12: 🔥🔥 One research paper is accepted by SIGKDD Explorations: Feature Selection for Fault Detection and Prediction based on Event Log Analysis
- 2021.10: 🔥🔥 One research paper is accepted by Clinical Trials: 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 the following 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 Dependable and Secure Computing (TDSC)
- IEEE Transactions on Emerging Topics in Computational Intelligence (TETCI)
- IEEE Transactions on Cloud Computing (TCC)
- IEEE Internet of Things Journal (IoT)
- IEEE Intelligent Systems
- Pattern Recognition
- Computer Networks
- AI Communications
Measurement(bad review experience)
- PC Member or Reviewer for the following conferences:
- ECMLPKDD
- KDD
- ICLR
- AAAI
- CVPR
Contact
WeChat: DigitalTwinNL Email: z.li(at)liacs.leidenuniv.nl