Weijia Wang

I am a PhD student in the PolSys team of LIP6, Sorbonne Université, and in the Sierra team at Inria Paris. I am supervised by Simone Naldi at XLIM, Mohab Safey El Din at LIP6, and Adrien Taylor at Inria Paris.

My research centers on designing computer algebra-based algorithms to automate the convergence analysis of first-order optimization algorithms. At present, these analyses rely heavily on ad hoc methodologies and intricate, manually written proofs with significant computational content. By enabling automated, computer-assisted proofs, this work has the potential to reshape the field fundamentally.

You can contact me via email at weijia.wang[at]{lip6.fr,sorbonne-universite.fr,inria.fr}, or find me at 26-00/315 on the Jussieu campus.

Teaching

  • Spring 2025: LU3IN024 - Introduction à la cryptologie (tutorials, practical work)
    TD 1 / TD 2 / TD 3 / TD 4 / TD 5 / TD 6 / TD 7 / TD 8 / TD 9 / TD 10 / TD 11
  • Fall 2024: LU2IN002 - Introduction à la programmation objet (practical work)

Publications

  • Consistency of automated market makers, Vincent Danos, Weijia Wang
    4th International Conference on Blockchain Economics, Security and Protocols (Tokenomics 2022). hal / doi / pdf

Preprints

  • Solving generic parametric linear matrix inequalities, Simone Naldi, Mohab Safey El Din, Adrien Taylor, Weijia Wang
    To appear in Proceedings of the 2025 International Symposium on Symbolic and Algebraic Computation (ISSAC 2025). hal / arXiv / pdf

Activities

  • Talk: Solving Generic Parametric Linear Matrix Inequalities. (upcoming)
    ISSAC 2025, Guanajuato, Mexico, 28 July-1 August 2025. program
  • Talk: Solving Parametric Linear Matrix Inequalities and Convergence Analysis in Optimization. (upcoming)
    SIAM Conference on Applied Algebraic Geometry (AG25), Madison, Wisconsin, U.S., 7-11 July 2025. abstract
  • Poster: Résolution des inégalités matricielles linéaires paramétriques génériques et applications en optimisation.
    JNIFM 2025, Bordeaux, France, 24-27 March 2025