Guillaume Braun

About Me

I am a Ph.D. student at Inria Lille - Nord Europe in MODAL team since January 2020. I work under the direction of Christophe Biernacki and Hemant Tyagi on inverse problems on graphs (mostly clustering and matching problems). I’m particulary interested in iterative methods including model-based approaches like EM algorithms and model-free approaches like generalized power methods. I’m also familiar with spectral methods, perturbation theory and concentration inequalities.

Publications

• Minimax Optimal Clustering of Bipartite Graphs with a Generalized Power Method
  G. Braun and H. Tyagi [ArXiv]

• Seeded graph matching for the correlated Wigner model via the projected power method
  E. Araya, G. Braun and H. Tyagi [ArXiv]

• An iterative clustering algorithm for the Contextual StochasticBlock Model with optimality guarantees
  G. Braun, H. Tyagi, and C. Biernacki, to appear in ICML 2022. [ArXiv]

• Clustering multilayer graphs with missing nodes
  G. Braun, H. Tyagi, and C. Biernacki, 24th International Conference on Artificial Intelligence and Statistics, 2021, 2260-2268. [PMLR] [Code][Poster]

Talks

• An iterative clustering algorithm for the Contextual Stochastic Block Model with optimality guarantees
  MODAL Seminar, Lille (online), November 30th, 2021

• Clustering graphs with side information
  SPSR Workshop, Bucharest (online), November 19th, 2021

• Clustering multilayer graphs with missing nodes
  MODAL Seminar, Lille, November 2020 [Slides]

Services

• Conference Reviewers: AISTATS 2021, AISTATS 2022.

Education

My detailed CV is avaible here

Powered by Jekyll and Minimal Light theme.