Dr Lan Du is an Associate Professor at Monash University and is affiliated with the machine learning group in the Department of Data Science and AI. As a faculty member, he is responsible for teaching and research in the field of machine learning. Additionally, Dr Du served as the faculty's director of Postgraduate Studies, overseeing all the postgraduate programs offered by the Faculty of IT and chairing the faculty's graduate program committee.
Before joining Monash University, Dr. Du worked as a postdoctoral research fellow with Professor Mark Johnson at Macquarie University. During his time there, he conducted research in natural language processing and machine learning. Dr Du earned his PhD in computer science from the Australian National University, where he worked under the supervision of Professor Wray Buntine and Dr. Huidong Jin. His research focused on developing Bayesian nonparametric models for topic modelling.
His research interest lies in the joint area of machine/deep learning and natural language processing, which includes (but is not limited to)
- Active learning: both theories and applications
- Uncertainty Estimation for NLP and CV on either uni-modal or multimodal data
- Knowledge distillation in CV and NLP
- Multi-view/modal learning
- AI for Chemistry: Molecular property prediction, molecular generation, etc
- AI for Economics
Dr Du has a strong interest in cross-disciplinary research, which involves applying machine learning and artificial intelligence techniques to solve real-world problems in different disciplines. He believes that the key to unlocking the full potential of AI lies in bridging the gap between advanced AI research and its practical application.
He has collaborated with researchers in fields including Medicine, Chemistry and Marketing, to develop AI solutions to address complex problems in these areas. His research focuses on developing innovative machine learning algorithms and models that can analyze and make sense of complex data sets to improve decision-making, predict outcomes, and drive innovation in various fields. By working across disciplines, he hopes to create new opportunities for applying AI and machine learning in practical, real-world settings, and to help ensure that these technologies are used in a responsible, ethical, and beneficial way.
Information for future PhD candidates
As an associate professor in machine learning, Dr Du is always looking for talented and motivated PhD students, who are passionate about pursuing research in these areas. His research group focuses on developing innovative algorithms and models for analyzing complex data sets, particularly on ML and its applications. Students who join Dr. Du's research group can expect to work on cutting-edge research projects in machine learning, using state-of-the-art techniques and tools to develop novel solutions to real-world problems.
To succeed in Dr Du's research group, students should be self-motivated, creative, and dedicated to advancing the state of the art in machine learning. They should also be able to work independently and collaboratively and have a strong background in computer science, mathematics, or a related field.
The PhD candidates can be found via (click the hyperlink for more information)
If you are interested in pursuing your PhD at Monash, and if you think you are competitive enough (E.g., the WAM of your master's study is equivalent to H1 honour, and you have gained some research experience) to apply for the scholarship, then
- Read how to apply, and review your own eligibility
- Fill in the Expression of Interest (EOI) form.
Recent Research highlights
- 2026
- CVPR 2026
- DPL: Decoupled Prototype Learning for Enhancing Robustness of Vision-Language Transformers to Missing Modalities
- IEEE Transactions on Multimedia 2026
- Adaptive Decoupled Knowledge Distillation via Self-learning
- CVPR 2026
- 2025
- CVPR 2025
- CGMatch: A Different Perspective of Semi-supervised Learning
- ICML 2025
- Navigating Conflicting Views: Harnessing Trust for Learning
- ACM Multimedia 2025
- Enhancing Multi-view Open-set Learning via Ambiguity Uncertainty Calibration and View-wise Debiasing
- CVPR 2025