Tensor Decompositions for Interpretable Machine Learning On Temporal Graphs

**Tensor Decompositions for Interpretable Machine Learning on Temporal Graphs**: - Réf **ABG-131020** - Sujet de Thèse - 14/04/2025 - Autre financement public - LISIC EA 4491, Université du Littoral Côte d'Opale - Lieu de travail- Saint-Omer - Les Hauts de France - France - Intitulé du sujet- Tensor Decompositions for Interpretable Machine Learning on Temporal Graphs - Champs scientifiques- Informatique - Science de la donnée (stockage, sécurité, mesure, analyse) - Mots clés- Machine Learning, Tensor decompositions, temporal graphs, anomaly detection, change point detection, wikipedia, industry 4.0 **Description du sujet**: Many real-world systems — such as industrial IoT networks, online marketplaces, and social platforms — produce data that naturally forms temporal graphs, where each interaction represents: who interacted with whom, and when. These graphs offer a powerful way to capture the evolving structure of complex systems and are key to detecting critical phenomena like fraud, cyberattacks, or the spread of misinformation. However, building machine learning tools for temporal graphs remains a major challenge, as common notions of similarity — essential for tasks like classification, prediction, or anomaly detection — are not easily defined in this context. This PhD project tackles that challenge by developing interpretable methods to compare temporal graphs. The core idea is to use tensor decompositions, which naturally represent temporal graphs as three-dimensional arrays and break them down into simpler/elementary building blocks. The tensor decomposition then opens door to compare temporal graphs by identifying common and different "atoms" across temporal graph instances. This PhD project aims to lay the foundation for interpretable machine learning on temporal graphs by developing similarity metrics that reveal meaningful patterns across time-evolving systems. To achieve this, we pursue three main goals: Goal 1: A new tensor decomposition for temporal graphs. We will develop a decomposition method tailored to the unique challenges of temporal graphs — including their binary and sparse nature, multi-scale patterns, and varying node sets or time intervals. Existing approaches fall short in capturing these aspects jointly. Our method will extend recent advances in geometric and coupled decompositions to extract shared and distinct motifs across graphs. Goal 2: Similarity metrics, machine learning tasks, and toolbox. Using our decomposition, we will define interpretable metrics that compare temporal graphs based on their structure, dynamics, and scale. These metrics will power machine learning algorithms for clustering, segmentation, change detection, and prediction. All methods will be implemented in a unified Python toolbox. Goal 3: Applications to real-world data. We will validate our methods on two domains. On Wikipedia, we aim to uncover patterns behind successful collaborations by analyzing contributor interactions. In Industry 4.0 settings, we will detect abnormal behaviors in sensor networks to identify faults or cyberattacks. **Nature du financement**: - Autre financement public **Précisions sur le financement**: - PEPR eNSEMBLE Data2Laws **Présentation établissement et labo d'accueil**: - LISIC EA 4491, Université du Littoral Côte d'Opale **Site web**: **Intitulé du doctorat**: - Doctorat en Informatique **Pays d'obtention du doctorat**: - France **Etablissement délivrant le doctorat**: - Université du Littoral Côte d'Opale **Ecole doctorale**: - ECOLE DOCTORALE EN SCIENCES, TECHNOLOGIE ET SANTE Starting date : October 2025 Duration: 3 years