Mobirise Website Builder

About me

I am currently researcher at Halias Technologies, and attached researcher at ERIC laboratory.

I passed my PhD in January 2025 under the supervision of Stéphane Chrétien for Université Lyon 2, and Laurent Testard for Halias.

My PhD manuscrit is available here.

Research
interests:

Signature transform

The signature of a path is a mathematical tool that encodes complex sequential data, such as time series or trajectories, into a sequence of iterated integrals. This representation captures the essential geometric and temporal structure of the data in a coordinate-free way.
Signatures provide a universal, compact, and theoretically rich framework for learning from streams of information, bridging ideas from rough path theory, stochastic analysis, machine learning and algebra.

Topological Data Analysis (TDA)

Topological Data Analysis studies the “shape” of data. Instead of focusing on local coordinates or pairwise distances, TDA extracts global and robust features that remain invariant under smooth deformations.
Through tools like persistent homology, TDA reveals structural patterns hidden in high-dimensional or noisy data, and it can complement or enhance standard statistical and geometric methods.

Causality

Causality aims to go beyond correlation: it seeks to understand how and why changes in one variable affect another.
Causal inference combines statistical reasoning, graphical models, and experimental design to uncover cause-effect relationships from data.
My current work focuses on applying concepts from pure mathematics, such as the theory of signatures, to explain and/or infer causal relationships.

‌

Free AI Website Builder