Hi! I’m Ting Gao, a Ph.D. candidate at Delft University of Technology working on machine learning for decision-making under uncertainty. My supervisors are Serge Hoogendoorn, Elvin Isufi, and Winnie Daamen.

My PhD started with solving real-world transport problems where data were noisy, sparse, and biased. Delivering models to Dutch municipalities taught me what makes machine learning work in practice: knowing the data deeply, making the right assumptions about the problem, and building those assumptions into the model itself.

But every new problem meant designing a new abstraction by hand. That pushed me toward a different question: can a model learn the right abstraction for a whole class of problems, instead of having it designed in for each one? Answering it took my work in a more theoretical direction, toward reinforcement learning, including work combining flow matching with policy optimization (NeurIPS 2026). At NAVER LABS Europe, I work with Jean-Marc Andreoli and Jean-Michel Renders on online stochastic combinatorial optimization, studying how to transfer what planning can do into reactive policies that act in real time.

Research interests: reinforcement learning, planning, generative models, combinatorial optimization, and graph learning

News

• 2026.10:  Received the Distinguished Award at NAVER LABS Europe Intern Day for my internship research on online stochastic optimization.
• 2026.10:  My personal website is online!
• 2026.09:  FMER is accepted at NeurIPS 2026.
• 2026.07:  I joined the Action group at NAVER LABS Europe as a research intern.
• 2026.06:  Q-Net is accepted at Transportation Research Part C.

Publications

Conference Publications

• Flow Matching Policy Optimization with Mirror Descent and Entropy Constraints [paper] [code] [blog]
  Ting Gao, Stavros Orfanoudakis, Nan Lin, Winnie Daamen, Serge Hoogendoorn, Elvin Isufi. NeurIPS 2026

Journal Publications

• Q-Net: Queue Length Estimation via Kalman-Based Neural Networks [paper]
  Ting Gao, Elvin Isufi, Winnie Daamen, Erik-Sander Smits, Serge Hoogendoorn. Transportation Research Part C: Emerging Technologies, 2026
• Bicycle Travel Time Estimation via Dual Graph-Based Neural Networks [paper] [code]
  Ting Gao, Winnie Daamen, Elvin Isufi, Serge P. Hoogendoorn. IEEE Transactions on Intelligent Transportation Systems, 2025
• Map-Matching for Cycling Travel Data in Urban Area [paper] [code]
  Ting Gao, Winnie Daamen, Panchamy Krishnakumari, Serge Hoogendoorn. IET Intelligent Transport Systems, 2024

Preprints

• Topological Kalman Filtering on Cell Complexes [paper]
  Chengen Liu, Rohan Money, Ting Gao, Mohammad Sabbaqi, Baltasar Beferull-Lozano, Elvin Isufi. arXiv 2026
• GT-CausIn: A Novel Causal-Based Insight for Traffic Prediction [paper]
  Ting Gao, Rodrigo Kappes Marques, Lei Yu. arXiv 2022

Patents

• Traffic Queue Estimator. Netherlands Patent Application N2041168 (Sep 2025)

Educations

• 2023.06 - now, Ph.D., Delft University of Technology, the Netherlands.
• 2020.09 - 2022.12, Diplôme d’Ingénieur (M.Sc. equivalent), CentraleSupélec, Université Paris-Saclay, France.
• 2020.09 - 2022.12, M.Sc. in Electrical and Electronics Engineering, Beihang University, China.
• 2016.09 - 2020.06, B.Sc. in Information and Computing Science, Beihang University, China.

Experience

• 2026.07 - now, Research Intern, NAVER LABS Europe, Grenoble, France.
• 2025.10, Guest Lecturer, PhD course Data, AI and Traffic Management, Delft University of Technology.
• 2025.07, Summer School, ELLIS Unit Cambridge Summer School on Probabilistic Machine Learning, University of Cambridge, UK.
• 2025.07, Summer School, London Geometry and Machine Learning (LOGML), Imperial College London, UK.
• 2025.06, Invited Speaker, Graph&Data@TUDelft seminar, Delft University of Technology.
• 2022.04 - 2023.05, Machine Learning Intern (Drug Discovery), StoneWise, Beijing, China.
• 2020.11 - 2021.04, Graduate Teaching Assistant, Convergence, Integration, Probability and Partial Differential Equations, CentraleSupélec, France.

Honors and Awards

• 2026 Distinguished Award, NAVER LABS Europe Intern Day.
• 2020 Outstanding Graduate of Beihang University.
• 2019 Chinese National Scholarship (top 0.2% nationwide).
• 2018 Chinese National Scholarship (top 0.2% nationwide).

Miscellany

• Languages: Chinese (native), French (C2), English (C1).
• Art: Outside research, I enjoy making art. You can find some of my artwork here.