Dear colleagues,
This is an announcement for the upcoming MALT seminar of Luc Brogat-Motte
Title
Provably Efficient Supervised Learning Algorithms for Non‑i.i.d. and Structured Data
Abstract
Luc’s research focuses on designing supervised machine‑learning algorithms that are both efficient and reliable. These methods come with formal performance guarantees, are released as open‑source software, and are validated on synthetic benchmarks as well as on real‑world datasets. In his doctoral work he developed structured‑prediction methods with convergence‑rate results and computational‑complexity bounds, demonstrating theoretical advantages and achieving significant improvements over state‑of‑the‑art approaches for applications such as metabolite identification. During his post‑doctoral research he shifted to learning from time‑dependent data, combining statistical‑learning theory with stochastic calculus. He developed estimation methods for continuous‑time, nonlinear stochastic dynamical systems and focused on stochastic differential‑equation estimation and safe learning for controlled systems under uncertainty, with applications in robotics and finance.
We hope to see many of you there!
Best regards,
MALT seminar organizing team