About me

I am a Research Fellow at the Division of Mathematical Sciences, Nanyang Technological University, working with Prof. Wang Zhongjian. Before joining NTU, I was an FCAI Postdoctoral Researcher at the Department of Computer Science, University of Helsinki and the Finnish Center for Artificial Intelligence, where I worked with Prof. Kai Puolamäki and Prof. Arto Klami. I obtained my PhD from the University of Lorraine (France) under the supervision of Prof. Hoai-An Le Thi and Prof. Hoai Minh Le. I defended my doctoral thesis in October 2022 before a jury composed of Prof. Anne Boyer, Prof. Stéphane Canu, Prof. Jalal Fadili, Prof. Yann Guermeur, Prof. Emilio Carrizosa, and Reader Xuan Vinh Doan. I earned a Master’s degree in Applied Mathematics with highest honors from Sorbonne Paris Nord University.

Research interests

- Nonconvex stochastic and Riemannian optimization.
- Geometric methods for machine learning and statistics.
- Sampling beyond log-concavity.
- Diffusion, flow, and PDE-based generative modeling.

News

[5.2026] The paper DC-LA with Wang Zhongjian was accepted to ICML. Both of us will be in Seoul to present our work. I was also recognized as a Silver Reviewer.
[3.2026] Invited talks at Inria Lyon (France) and City University of Hong Kong.
[12.2025] Had a good time attending NeurIPS in San Diego, California. Met some old friends.
[12.2025] I will be serving as a Reviewer for ICML 2026.
[10.2025] I was recognized as a Top Reviewer (Main track) for NeurIPS 2025.
[9.2025] Marcelo and I are looking for a PhD student for the project “Advances in generalized Bayesian inference via differential-geometric methods” (deadline Oct. 22). The project is funded by the Research Council of Finland and is further related to ELLIS. Apply here.
[8.2025] I will be serving as a Reviewer for ICLR 2026.
[7.2025] The paper GRADSTOP: Early Stopping of Gradient Descent via Posterior Sampling with Arash Jamshidi, Lauri Seppäläinen, Katsiaryna Haitsiukevich, Anton Björklund and Kai Puolamäki has been accepted for publication at the 28th European Conference on Artificial Intelligence (ECAI-2025). Congrats EDA @ Helsinki!
[7.2025] I will be joining the School of Physical and Mathematical Sciences, Nanyang Technological University (NTU Singapore) as a Research Fellow.
[5.2025] The paper Geodesic Slice Sampler for Multimodal Distributions with Strong Curvature with Bernardo Williams, Hanlin Yu, Georgios Arvanitidis, Arto Klami was accepted to UAI 2025. Congrats Monge team!
[4.2025] I will be in Singapore to attend ICLR and present our work.
[4.2025] I will be serving as a Reviewer for NeurIPS 2025.
[2.2025] The paper Stochastic variance-reduced Gaussian variational inference on the Bures-Wasserstein manifold with Hanlin Yu, Bernardo Williams, Marcelo Hartmann, Arto Klami was accepted to ICLR 2025.
[12.2024] I am in Vancouver, Canada to attend NeurIPS and present our work.
[12.2024] I will be serving as a Reviewer for ICML 2025.
[10.2024] I joined the Transaction of Machine Learning Research (TMLR) as a Reviewer.
[10.2024] I will be serving as a Reviewer for ECML PKDD 2024 Demo Track.
[9.2024] The paper Non-geodesically-convex optimization in the Wasserstein space with Hanlin Yu, Bernardo Williams, Petrus Mikkola, Marcelo Hartmann, Kai Puolamäki, Arto Klami was accepted to NeurIPS 2024.
[9.2024] I will be serving as a Reviewer for ICLR 2025.
[9.2024] I will be serving as a Reviewer for AISTATS 2025.
[8.2024] I reviewed for the 16th Asian Conference on Machine Learning (ACML-2024).
[5.2024] I am in Valencia, Spain to attend AISTATS 2024 and present our work.
[2.2024] The paper Markov chain stochastic DCA and applications in deep learning with PDEs regularization with Hoai An Le Thi and Hoai Minh Le was published at Neural Networks.
[1.2024] The paper Error bounds for any regression model using Gaussian processes with gradient information with Rafael Savvides and Kai Puolamäki was accepted to AISTATS 2024.
[10.2023] I will be serving as a Reviewer for AISTATS 2024.
[2.2023] I joined the Department of Computer Science, University of Helsinki, Finland as a postdoc.
[10.2022] I successfully defended my PhD thesis!
[10.2022] The paper "Online stochastic DCA with applications to principal component analysis" with Hoai An Le Thi and Tao Pham Dinh was accepted to IEEE Transactions on Neural Networks and Learning Systems.
[7.2022] The paper "Stochastic DCA with Variance Reduction and Applications in Machine Learning" with Hoai An Le Thi, Hoai Minh Le, Tao Pham Dinh was published at JMLR.
[6.2022] The paper "Stochastic difference-of-convex-functions algorithms for nonconvex programming" with Hoai An Le Thi, Van Ngai Huynh, Tao Pham Dinh was published at SIAM Journal on Optimization.
[4.2022] The paper "A DC programming approach for solving a centralized group key management problem" with Hoai An Le Thi and Thi Tuyet Trinh Nguyen was accepted to Journal of Combinatorial Optimization.

Check out my Google scholar (full list), DBLP, and OpenReview for my publications.
Both my Erdős number and Einstein number are 5.

Selected Publications

Non-geodesically-convex optimization in the Wasserstein space.
Luu, H. P. H., Yu, H., Williams, B., Mikkola, P., Hartmann, M., Puolamäki, K., & Klami, A. (2024). Advances in Neural Information Processing Systems (NeurIPS).

DC-LA: Difference-of-convex Langevin Algorithm.
Luu, H. P. H. & Wang, Z. (2026). International Conference on Machine Learning (ICML).

Stochastic variance-reduced Gaussian variational inference on the Bures-Wasserstein manifold.
Luu, H. P. H., Yu, H., Williams, B., Hartmann, M., & Klami, A. (2025). International Conference on Learning Representations (ICLR).

Stochastic DCA with variance reduction and applications in machine learning.
Le Thi, H. A., Luu, H. P. H., Le, H. M., & Pham Dinh, T. (2022). Journal of Machine Learning Research (JMLR), 23(206), 1–44.

Stochastic difference-of-convex-functions algorithms for nonconvex programming.
Le Thi, H. A., Huynh, V. N., Pham Dinh, T., & Luu, H. P. H. (2022). SIAM Journal on Optimization, 32(3), 2263–2293.

Error bounds for any regression model using Gaussian processes with gradient information.
Savvides, R., Luu, H. P. H., & Puolamäki, K. (2024). In Proceedings of The 27th International Conference on Artificial Intelligence and Statistics (AISTATS), PMLR 238:397–405.

Geodesic slice sampler for multimodal distributions with strong curvature.
Williams, B., Yu, H., Luu, H. P. H., Arvanitidis, G., & Klami, A. (2025). In Proceedings of the 41st Conference on Uncertainty in Artificial Intelligence (UAI). Proceedings of Machine Learning Research.

Markov chain stochastic DCA and applications in deep learning with PDEs regularization.
Luu, H. P. H., Le, H. M., & Le Thi, H. A. (2024). Neural Networks, 170, 149–166.

Online stochastic DCA with applications to principal component analysis.
Le Thi, H. A., Luu, H. P. H., & Pham Dinh, T. (2022). IEEE Transactions on Neural Networks and Learning Systems, 35(5), 7035–7047.

Contact

Dr. LUU Hoang Phuc Hau
School of Physical and Mathematical Sciences (SPMS)
21 Nanyang Link 637371 Singapore
Email: hoangph.luu@ntu.edu.sg (or hoang-phuc-hau.luu@helsinki.fi)