news

I successfully defended my PhD thesis First-Order Methods for Geodesically Convex Optimization on Riemannian Manifolds with distinction! Many thanks to my committee: Sebastian Pokutta, David Martínez-Rubio, Martin Skutella, Suvrit Sra, and Tobias Breiten.

New preprint on arXiv: Restart and Adaptive Acceleration in Stochastic Gradient Methods

The Agentic Researcher was accepted as an oral presentation at the ICML 2026 Workshop AI4Research!

New preprint on arXiv: Curvature-Dependent Lower Bounds for Frank-Wolfe

Two papers accepted at ICML 2026:

• Lower Bounds for Frank-Wolfe on Strongly Convex Sets
• From Associations to Activations: Comparing Behavioral and Hidden-State Semantic Geometry in LLMs

Teaching this semester: High-Dimensional Optimization and Learning at TU Berlin. Thursdays, 14:00–16:00, MAR 4.064. Course material on ISIS.

Gave a talk on “Lower Bounds for Frank-Wolfe on Strongly Convex Sets” at the SIERRA Seminar at INRIA Paris.

New preprint on arXiv: The Agentic Researcher: A Practical Guide to AI-Assisted Research in Mathematics and Machine Learning

I’m visiting Alexandre d’Aspremont at INRIA Paris from February to April 2026.

Two new preprints on arXiv:

• Lower Bounds for Frank-Wolfe on Strongly Convex Sets
• From Associations to Activations: Comparing Behavioral and Hidden-State Semantic Geometry in LLMs