Guillaume Braun
Scholar

Guillaume Braun

Google Scholar ID: lzebyjAAAAAJ
Riken
Inverse problems on graphsLow dimensional models for high dimensional data
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
59
 
H-index
4
 
i10-index
2
 
Publications
12
 
Co-authors
0
 
Contact
GitHubOpen ↗
Publications
5 items
Overfitting of Spectral Gradient Descent: How Matrix Geometry shapes Generalization and Implicit Bias
2026
Cited
0
Spectral Gradient Descent Mitigates Anisotropy-Driven Misalignment: A Case Study in Phase Retrieval
2026
Cited
0
Neuron Block Dynamics for XOR Classification with Zero-Margin
2026
Cited
0
Fast Escape, Slow Convergence: Learning Dynamics of Phase Retrieval under Power-Law Data
2025
Cited
0
Learning a Single Index Model from Anisotropic Data with vanilla Stochastic Gradient Descent
2025
Cited
0
Resume
Academic Achievements
  • - Publications:
  • * Learning a Single Index Model from Anisotropic Data with Vanilla Stochastic Gradient Descent, AISTATS (2025)
  • * VEC-SBM: Optimal Community Detection with Vectorial Edge Covariates, AISTATS (2024)
  • * Strong Consistency Guarantees for Clustering High-Dimensional Bipartite Graphs with the Spectral Method, Electronic Journal of Statistics (2024)
  • * Minimax Optimal Clustering of Bipartite Graphs with a Generalized Power Method, Information and Inference: A Journal of the IMA (2023)
  • * Seeded graph matching for the correlated Wigner model via the projected power method, Journal of Machine Learning Research (2024)
  • * An iterative clustering algorithm for the Contextual Stochastic Block Model with optimality guarantees, ICML (2022)
  • * Clustering multilayer graphs with missing nodes, AISTATS (2021)
  • - Talks:
  • * Journées MAS 2024, Poitiers
  • * SIAM Workshop on Network Science 2022
  • * ICML 2022
  • * MODAL Seminar, Lille (online), November 30th, 2021
  • * SPSR Workshop, Bucharest (online), November 19th, 2021
  • * MODAL Seminar, Lille, November 2020
  • - Reviewing:
  • * Conferences: AISTATS 2021-2025, ICML 2022
  • * Journals: Journal of Machine Learning Research, Annals of Statistics
Research Experience
  • - April 2024 to present: Postdoctoral Researcher, High-Dimensional Causal Analysis Team, RIKEN AIP
  • - 2023 to 2024: Postdoctoral Researcher, Imperfect Information Learning Team
Education
  • - Ph.D. Degree: Inria Lille - Nord Europe, MODAL team, supervised by Christophe Biernacki and Hemant Tyagi, focusing on clustering and matching problems on graphs (specific time not provided)
Background
  • - Research Interests: Theoretical machine learning and high-dimensional statistics
  • - Professional Field: Clustering and matching problems on graphs
  • - Brief Introduction: Since April 2024, he has been a postdoctoral researcher in the High-Dimensional Causal Analysis Team at RIKEN AIP, under the supervision of Masaaki Imaizumi.