Publications

Artificial Intelligence

Pattern detection and shape analysis for physiological timeseries

Time series are prevalent in biomedical applications where they frequently display recurring or abnormal patterns that hold significant information for statistical analysis. A notable example is the heartbeat in electrocardiograms, a recurring…

Thibaut Germain

Artificial Intelligence

Structured Loss for Deep Change-Point Detection

Change-point detection, which is finding abrupt changes in the behaviour of a signal, is a crucial step in numerous data processing pipelines. There are many available algorithms in the literature; choosing one and calibrating it is complex and time…

Simon Blotas, Charles Truong

32nd European Signal Processing Conference (EUSIPCO) - 2024

Computer Science

An Efficient Algorithm for Exact Segmentation of Large Compositional and Categorical Time Series

Change‐point detection, also known as signal segmentation, is an essential preprocessing step in many applications, ranging from industrial monitoring to bioinformatics. In short, it consists in finding the temporal boundaries of homogeneous regimes…

Charles Truong, Vincent Runge

Stat - 2024

Computer Science

Une variante pondérée de K-means adaptée aux données textuelles

This article proposes a weighted variant of the K-means algorithm, specifically adapted to textual data. The algorithm aims to improve clustering performance and interpretability by assigning weights to the most discriminating features for each…

Amine Ferdjaoui, Séverine Affeldt, Mohamed Nadif

29èmes Rencontres de la Société Francophone de Classification - 2024

Computer Science

Convolutional Sparse Coding for Time Series Via a l0 Penalty: An Efficient Algorithm With Statistical Guarantees

Identifying characteristic patterns in time series, such as heartbeats or brain responses to a stimulus, is critical to understanding the physical or physiological phenomena monitored with sensors. Convolutional sparse coding (CSC) methods, which…

Charles Truong, Thomas Moreau

Statistical Analysis and Data Mining - 2024

Mathematics

Free Boundary Problems in Fluid Dynamics

Albert Ai, Thomas Alazard, Mihaela Ifrim, Daniel Tataru

Atmospheric and Oceanic Physics

Beyond the Sea Surface : Exploring Ocean Wave Measurements From Multiple Perspectives

This thesis explores ocean wave measurements from multiple perspectives, extending the analysis beyond their visual manifestation on the sea surface. Existing limitations in measurement technologies hinder a full understanding of aspects such as…

Daniel Santiago Pelaez Zapata

Image Processing

Detecting Forged Sentinel-2 Images Through Parallax-Based Cloud Analysis

The availability and significance of satellite imagery in our world is continuously growing. Satellite images now play a crucial role in various applications such as weather forecasting, greenhouse gas monitor- ing, agricultural crop health…

Matthieu Serfaty, Quentin Bammey, Tina Nikoukhah, Rafael Grompone von Gioi, Carlo de Franchis

European Conference on Computer Vision (ECCV) 2024 - 2024

Mathematics

Graph-Convolutional Networks: Named Entity Recognition and Large Language Model Embedding in Document Clustering

Recent advances in machine learning, particularly Large Language Models (LLMs) such as BERT and GPT, provide rich contextual embeddings that improve text representation. However, current document clustering approaches often ignore the deeper…

Imed Keraghel, Mohamed Nadif

Computer Science

Clustering de documents avec des plongements LLMs : évaluation de la réduction de dimension

Le clustering de documents est crucial pour la gestion et l'analyse d'importants volumes de données textuelles. Les progrès récents dans les grands modèles de langage (LLMs), comme GPT, ont transformé la manière dont les textes peuvent être analysés…

Imed Keraghel, Stanislas Morbieu, Mohamed Nadif

Les 29èmes Rencontres de la Société Francophone de Classification (SFC) - 2024