Antoine Groudiev

Computer Science Departement, École Normale Supérieure - PSL, Paris, France

prof_pic.jpg

I am a second year Master’s student in Computer Science at the École Normale Supérieure, Paris, France. I am currently enrolled in the MVA program (Mathematics, Vision, Learning) at ENS Paris-Saclay. My research interests include robotics, optimization, and machine learning.

From March to August 2025, I visited the Computational Robotics Group under the supervision of Heng Yang at Harvard University. Prior to that, I had the chance to do a research project on optimization for robotics in the Willow team at Inria in Paris, advised by Frederike Dümbgen and Justin Carpentier. In 2024, I worked on explainable AI at LIG in Grenoble, France.

selected publications

  1. RSS Workshop
    cuADMM: GPU-Accelerated First-Order Optimization for Large-Scale Multi-Block Semidefinite Programs
    Antoine Groudiev, Shucheng Kang, and Heng Yang
    In 2025 RSS Workshop on Fast Motion Planning and Control in the Era of Parallelism, 2025
    PDF Code
  2. arXiv
    KernelSOS for Global Sampling-Based Optimal Control and Estimation via Semidefinite Programming
    Antoine Groudiev, Fabian Schramm, Éloïse Berthier, Justin Carpentier, and Frederike Dümbgen
    arXiv preprint, 2025
    Best Poster Award PDF Poster
  3. PW25
    Extending Layer-wise Relevance Propagation in Neural Networks using Semiring Annotations
    Antoine Groudiev, Arkaprava Saha, and Silviu Maniu
    In Proceedings of the ProvenanceWeek 2025, 2025
    PDF Code
  4. arXiv
    Factorization-free Orthogonal Projection onto the Positive Semidefinite Cone with Composite Polynomial Filtering
    Shucheng Kang, Haoyu Han, Antoine Groudiev, and Heng Yang
    arXiv preprint, 2025
    PDF Code

news

Jul 24, 2025 Our poster on KernelSOS received the Best Poster Award at the 2025 IEEE RAS Technical Committee on Model-Based Optimization for Robotics Poster Event!
Jul 23, 2025 New preprint on KernelSOS for Global Sampling-Based Optimal Control and Estimation via Semidefinite Programming.
Jul 14, 2025 New preprint on Factorization-free Orthogonal Projection onto the Positive Semidefinite Cone with Composite Polynomial Filtering.
Jun 04, 2025 cuADMM: GPU-Accelerated First-Order Optimization for Large-Scale Multi-Block Semidefinite Programs accepted to the 2025 RSS Workshop on Fast Motion Planning and Control in the Era of Parallelism.
May 16, 2025 I presented Globally-Optimal Control and Estimation via Kernel Sums of Squares in the Robot Locomotion Group at MIT.