Photo

About Me

Welcome to my academic homepage! I am a PhD student at LIRIS specializing in computer science and Artificial Intelligence. My research focuses on graph representation learning and its applications.

Contact Information

Email: elouan.vincent@insa-lyon.fr

Team: DM2L@LIRIS

Research Interests

Graph machine learning

Developing novel graph neural network architectures and representation learning methods for complex structured data. I am currently working on a new interpretable graph kernel by extracting subgraph of interest with resepect to the predictive task.

Explainable AI (xAI)

Creating interpretable machine learning models and post-hoc explanation methods to understand AI decision-making processes. I investigate techniques for making black-box models more transparent, developing visualization tools for model explanations, and ensuring AI systems are trustworthy and accountable in critical applications.

Anomaly Detection

Researching unsupervised and semi-supervised methods for detecting outliers and anomalies in complex datasets. My current work explores the intersection of diffusion models and anomaly detection, focusing on creating more interpretable and robust detection systems for applications in cybersecurity, fraud detection, and system monitoring.

Latest News

Aug 2025

New Paper Accepted

Our paper "Diffusion for Explainable Unsupervised Anomaly Detection" has been accepted to DSAA 2025. This work explores the use of diffusion processes for enhancing the interpretability of unsupervised anomaly detection methods.

Jul 2025

Student Volunteer at COLT 2025

Excited to volunteer at the Conference on Learning Theory (COLT 2025)! Looking forward to contributing to the machine learning theory community and learning from leading researchers in the field.