My CV
Publications
Talks
Teaching
Numerics
Main page

Contact

clement.royer@lamsade.dauphine.fr
Clément Royer
Université Paris Dauphine-PSL
Place du Maréchal de Lattre de Tassigny
75016 Paris
FRANCE
I am also on Google Scholar and Github.

Clément W. Royer

Associate professor, Université Paris Dauphine-PSL


Pour la version française de cette page, cliquez ici.

Latest news

  • Accepted paper (2024/10) The paper Complexity analysis of regularization methods for implicitly constrained least squares, co-authored with Akwum Onwunta, has appeared in Journal of Scientific Computing.

  • Teaching (2024/09) The Fall semester has started! I will be teaching several courses around linear algebra and applications to data science (bachelor level) and optimization for machine learning (master level). Details can be found here.

  • Accepted paper (2024/08) The paper Expected decrease for derivative-free algorithms using random subspaces, co-authored with Warren Hare and Lindon Roberts, has appeared online in Mathematics of Computation.

  • Accepted paper (2024/08) The paper Full-low evaluation methods for bound and linearly constrained derivative-free optimization., co-authored with Oumaima Sohab and Luis Nunes Vicente, has appeared online in Computational Optimization and Applications.

  • Research stay (2024/07) I visited Albert Berahas at the University of Michigan (Ann Arbor, MI, USA) during the last week of July. Thanks to Albert and the IOE department for being such great hosts. Stay tuned for new work from us!

  • Conference (2024/07) I attended the 25th International Symposium on Mathematical Programming in Montréal (QC, Canada), on July 21-26. A great event to catch up with the community! My presentation (in the derivative-free optimization stream) can be found here.

  • Conference (2024/06) I attended the second Derivative-Free Optimization Symposium, held at the University of Padova (Italy) on June 24-28. I gave a talk based on joint works with Warren Hare, Gabriel Jarry-Bolduc, as well as my PhD student Sébastien Kerleau (who also gave a talk at the conference). My slides are available here. Huge thanks to the organizing committee for this great event!

  • Conference (2024/06) I gave a talk at the Workshop on Bayesian optimization and related topics, organized by Julien Bect, Céline Helbert and Delphine Sinoquet. I presented my results on randomized derivative-free optimization algorithms, slides can be found here.

  • Visitor (2024/06) Lindon Roberts (University of Sydney) visited me in Dauphine from June 13 to June 21.

  • Conference (2024/05) I organized a session on negative eigenvalues and nonconvex optimization during the 2024 SIAM Applied Linear Algebra conference. My kick-off slides for this session can be found here. Thanks to the other speakers Iskander Legheraba, Sadok Jerad and Irène Waldspurger for accepting my invitation!

  • Grant (2024/04) I am the recipient of a CNRS IEA (International Emerging Actions) grant! My grant BONUS (Blackbox Optimization with a Novel Use of Subspaces), joint with Lindon Roberts from the University of Sydney, will allow us to investigate novel use of subspace techniques in derivative-free optimization. See here for details.

  • Conference (2024/03) I attended the 2024 edition of the SMAI-MODE Days in Lyon, France, where I gave a talk on derivative-free optimization over subspaces.

  • Conference (2024/03) I attended the 25th annual conference of ROADEF, the French Operations Research society, in Amiens. I was a member of the student paper prize (check out the two recipient papers by Margot Herin and Victor Spitzer!) and also gave a talk on an ongoing research project with colleagues at Dauphine (learn more here).

  • New preprint (2024/03) In Revisiting theoretical guarantees of direct-search methods, we survey the theory around direct-search algorithms using sometimes novel proof techniques! Joint work with Kwassi Joseph Dzahini from Argonne National Laboratory (USA) as well as Francesco Rinaldi and Damiano Zeffiro from Università Degli Studi di Padova (Italy).

  • New preprint (2024/02) In Riemannian trust-region methods for strict saddle functions with complexity guarantees, we show that trust-region schemes enjoy better complexity guarantees on strict saddle functions than on generic nonconvex ones. A long-time effort led by my postdoc Florentin Goyens!

    Research interests

    My research essentially revolves around the field of numerical optimization and its applications, particularly in complex systems and data science. In addition to optimization, my work relies heavily on linear algebra techniques and randomized methods.
    Lately I have been interested in analyzing certain classes of nonconvex optimization problems and developing efficient algorithms with complexity guarantees for solving those problems. I am also highly interested in derivative-free optimization and its applications to solving simulation-based problems.

    Short bio

  • September 1st, 2019: Joined Université Paris Dauphine-PSL and LAMSADE as a maître de conférences (equivalent to associate professor, with tenure granted on September 1st, 2020).

  • November 2016-August 2019: Postdoctoral research associate within the Wisconsin Institute for Discovery, in Madison, Wisconsin (USA). I was fortunate to work in the group of Stephen J. Wright.

  • November 4, 2016: PhD in applied mathematics from the university of Toulouse, delivered by the Université Toulouse III Paul Sabatier. Prepared at the Institut de Recherche en Informatique de Toulouse, under the joint supervision of Serge Gratton and Luís Nunes Vicente.

  • For more information, you can have a look at my vitae in a short or extended format.

    Materials on this page are available under Creative Commons CC BY-NC 4.0 license.