Biography

I am an AI Fellow at the Paris School of AI (PRAIRIE-PSAI) at Université Paris Sciences et Lettres since October 2025. My research activities take place within the LAMSADE (Laboratoire d’analyse et modélisation de systèmes pour l’aide à la décision) at Université Paris Dauphine-PSL. From 2024 to 2025, I was a postdoctoral researcher at the L2S (Laboratory of Signals and Systems), CentraleSupélec, CNRS, Université Paris-Saclay, and from 2021 to 2024, I was a postdoctoral researcher within the TAU Team at INRIA, LISN, Université Paris-Saclay. My research focuses on causal learning, matrix and tensor decomposition and related topics in optimization. I obtained my PhD in May 2021 at Université catholique de Louvain, Belgium, under the supervision of Pierre-Antoine Absil and Kyle A. Gallivan. During my PhD I worked on matrix and tensor completion, graph-based machine learning, and optimization on matrix manifolds.

Publications and Preprints

• A fast non-iterative method for learning latent variable causal models
  Shuyu Dong, Arthur Tenenhaus, Laurent Le Brusquet, Mohammed Nabil El Korso. HAL, 2026.

• Learning Measurement Models via Subspace Identification and Clustering
  Shuyu Dong, Arthur Tenenhaus, Laurent Le Brusquet, Mohammed Nabil El Korso. IEEE Statistical Signal Processing Workshop (SSP), 2025, pp. 211-215.
  [PDF]

• DCILP: A distributed approach for large-scale causal structure learning
  Shuyu Dong, Michèle Sebag, Kento Uemura, Akito Fujii, Shuang Chang, Yusuke Koyanagi, and Koji Maruhashi. The 39th Annual AAAI Conference on Artificial Intelligence (AAAI-25), 2025.
  [PDF] [arXiv]

• Learning Large Causal Structures from Inverse Covariance Matrix via Matrix Decomposition
  Shuyu Dong, Kento Uemura, Akito Fujii, Shuang Chang, Yusuke Koyanagi, Koji Maruhashi, and Michèle Sebag. arXiv:2211.14221, 2024.

• From graphs to DAGs: a low-complexity model and a scalable algorithm
  Shuyu Dong and Michèle Sebag. European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD), 2022
  [arXiv]

• On the analysis of optimization with fixed-rank matrices: a quotient geometric view
  Shuyu Dong, Bin Gao, Wen Huang, and Kyle A. Gallivan. arXiv:2203.06765, 2024.

• New Riemannian preconditioned algorithms for tensor completion via polyadic decomposition
  Shuyu Dong, Bin Gao, Yu Guan, and François Glineur. SIAM Journal on Matrix Analysis and Applications 43 (2) (2022), 840-866.
  [arXiv]

• Alternating minimization algorithms for graph regularized tensor completion
  Yu Guan, Shuyu Dong, Bin Gao, P.-A. Absil, and François Glineur. arXiv:2008.12876, 2023.

• Riemannian gradient descent methods for graph-regularized matrix completion
  Shuyu Dong, P.-A. Absil, and Kyle A. Gallivan. Linear Algebra and its Applications 623 (2021), 193-235.
  [PDF]

• Preconditioned conjugate gradient algorithms for graph regularized matrix completion
  Shuyu Dong, P.-A. Absil, and Kyle A. Gallivan. The 27th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN), 2019.

• Graph learning for regularized low-rank matrix completion
  Shuyu Dong, P.-A. Absil, and Kyle A. Gallivan. The 23rd International Symposium on Mathematical Theory of Networks and Systems (MTNS), 2018.

• Learning sparse models of diffusive graph signals
  Shuyu Dong, Dorina Thanou, P.-A. Absil, and Pascal Frossard. The 25th European Symposium on Artificial Neural Networks (ESANN), 2017.