Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : LISIC – Univ. Littoral Côte d’Opale
Durée : 3 years
Contact : esteban.bautista@univ-littoral.fr
Date limite de publication : 2025-06-01

Contexte :
Many complex systems — such as the Internet, transportation networks, and financial systems — produce data that naturally takes the form of temporal graphs, where each link between nodes is time-stamped. Temporal graphs allow us to model and analyze interactions over time, such as network traffic between computers, scheduled trips between stations, or transactions between bank accounts. A common challenge in working with such graphs is identifying subsets of the temporal graph that optimize certain properties, like density, cost, duration, etc. These tasks are essential for applications such as anomaly detection, cybersecurity, or route planning, but they typically involve NP-hard combinatorial problems, making them impractical to solve exactly.

This PhD project explores a new direction for tackling these problems using artificial intelligence. While heuristic methods exist, they often struggle to balance speed and accuracy in temporal graph settings. In contrast, recent advances show that AI models can be trained to solve combinatorial problems on static graphs efficiently, yet their potential remains largely unexplored in the temporal graph setting. This project aims to bridge that gap by developing AI-based methods that learn to solve combinatorial optimisation problems emerging on temporal graphs.

Sujet :
This PhD project aims to explore the potential of machine learning methods as a means to solve combinatorial optimization problems on temporal graphs. We target three specific goals:

Goal 1: End-to-end learning framework.
We aim to design a framework that trains neural models to directly map problem instances to solutions in temporal graphs. While such approaches exist for static graphs, our challenge is to extend them to the temporal setting by defining suitable loss functions and training strategies.

Goal 2: A novel filter-based architecture.
We plan to develop a neural architecture that treats optimization as a filtering task — discarding irrelevant links to isolate the optimal subgraph. Building on spectral methods and recent work in temporal graph signal processing, we will explore how filters can be effectively defined and learned in a frequency-structural domain.

Goal 3: High-impact applications.
We will validate our methods on two key applications:

– Anomaly localization: Many systems detect anomalies but fail to pinpoint their origin. We aim to learn to localize anomalies without relying on assumptions about their structure.

– Temporal graph exploration: In transportation networks, finding optimal exploration routes is NP-hard. Our goal is to develop practical AI-based methods that scale better than current approximations.

Profil du candidat :
We look for highly motivated candidates with relevant experience in computer science, graph algorithms, and/or
deep learning. Experience in Python programming and operations research will be a plus.

—— Application ——-

Interested candidates are invited to send an e-mail to

• esteban.bautista@univ-littoral.fr
• rym.guibadj@univ-littoral.fr
• gilles.roussel@univ-littoral.fr

while attaching the documents that can support their application:

• your resume;
• a cover letter;
• your transcripts from the last year of B.Sc to the last year of M.Sc. (if the latter is already available);
• two reference letters or the names and means of contact of two academic advisers.

Applications will be reviewed on a rolling basis until the position is filled.

Formation et compétences requises :

Adresse d’emploi :
LISIC Laboratory – Université du Littoral Côte d’Opale – Site Saint-Omer (Hautes de France), France

Document attaché : 202504141925_PhD_COTEG.pdf

Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : LISIC – Univ. Littoral Côte d’Opale
Durée : 3 years
Contact : esteban.bautista@univ-littoral.fr
Date limite de publication : 2025-06-01

Contexte :
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. Temporal 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. Such temporal graph “atoms” can be leveraged to identify what fundamental building blocks are common and different across temporal graph instances, resulting in a set of tools that not only enhance machine learning on temporal graphs but also offer insights into the underlying patterns driving complex systems.

Sujet :
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.

Profil du candidat :
We look for highly motivated candidates with relevant experience in computer science, data science, and graph machine learning. Experience in Python programming and signal processing will be a plus.

Interested candidates are invited to send an e-mail to
• esteban.bautista@univ-littoral.fr
• laurent.brisson@imt-atlantique.fr
• matthieu.puigt@univ-littoral.fr
while attaching the documents that can support their application:
• your resume;
• a cover letter;
• your transcripts from the last year of B.Sc to the last year of M.Sc. (if the latter is already available)
• two reference letters or the names and means of contact of two academic advisers.

Applications will be reviewed on a rolling basis until the position is filled.

Formation et compétences requises :

Adresse d’emploi :
LISIC Laboratory – Université du Littoral Côte d’Opale – Site Saint-Omer (Hautes-de-France), France

Document attaché : 202504141906_PhD_Data2Laws.pdf

Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : Centre INRIA de l’Université de Rennes
Durée : 3 ans
Contact : nicolas.gutowski@univ-angers.fr
Date limite de publication : 2025-05-15

Contexte :
The project involves three partners: Inria, Leria and DGA TT.

Inria is the National Research Institute for Digital Science and Technology. This center for scientific excellence is currently directing the French Digital Programs Agency and is on the frontline of digitalization in Europe while conducting world-class research covering a wide range of disciplines: computer science, mathematics, and simulation software. International and industrial collaborations, ground-breaking research, software development, artificial intelligence, quantum- and cyber technologies and deep tech startups are the DNA of the institute. Inria ranks 16th worldwide at the AI Research ranking while being the number one European institute for frontier research in digital sciences.

LERIA (Laboratory for Computer Science Research in Angers) is a dynamic research unit of the University of Angers, bringing together around 40 members, including 23 faculty researchers and a vibrant community of PhD students and collaborators. Its scientific focus lies at the intersection of Artificial Intelligence and Optimization, with research spanning from theoretical models for knowledge representation and reasoning to advanced algorithms for solving complex combinatorial problems. Bridging fundamental research and real-world applications, LERIA actively contributes to both academic and industrial innovation in intelligent computing.

DGA TT (French Procurement Agency – Land Techniques) is the French defense center dedicated to supporting and evaluating all land-based military systems. Its teams specialize in combat systems, ergonomics, weaponry, protection, robotics, and vehicle dynamics. With key sites in Bourges and Angers, the Angers facility focuses on the characterization and real-world testing of military vehicles, including their dynamic behavior and resistance to harsh environments. Spanning over 150 hectares, the site offers specialized tracks and state-of-the-art facilities to simulate extreme mechanical and climatic conditions – ensuring that equipment performs reliably under the most demanding scenarios.

Recruiting team: The recruited researcher will join a multi-disciplinary team involving established, full-time research scientists of all ages, MSc, PhD and postdocs (~30 ppl). The training programme intends to prepare candidates for scientific positions, either in academia or industry, by working in a research-intensive environment which fosters both scientific excellence (world-class researchers and over 65 prestigious ERC grants) and entrepreneurship (over 200 startups launched and a dedicated Inria Startup Studio). You will also have access to an extensive portfolio of training courses on digital science and technology, scientific programming or Artificial Intelligence. The candidate will have the opportunity to share his time in Rennes and Angers, through a flexible programme. He will be recruited by Inria in the Beaulieu Scientific Campus of University of Rennes, Bretagne (France), a medium town (~220.000 inhabitants) close from Paris and from the sea, with an intense student life (25% of the population). and Angers (France’s top city to live in, 2022) and Rennes (1st student city in France, 2024, and 8th in Europe for quality of life, 2019) both offer an exceptional quality of life, a rich history, vibrant culture, and a thriving economy.

Sujet :
Apply here : https://jobs.inria.fr/public/classic/fr/offres/2025-08786

This PhD project aims for an advance in the simulation and design of metamaterials – engineered structures with unique physical properties surpassing natural materials. It aims to develop automated finite element methods to enhance numerical modeling, focusing on precision and efficiency in simulations while enabling predictive design of adaptive systems. The research targets critical defense technologies, including stealth, cloaking, and wave manipulation, with applications spanning multispectral furtivity, dynamic resilience, and active wave modulation, relevant to both defense and civilian contexts.

This PhD project, titled “Artificial Intelligence for Optimizing the Mechanical Resilience of Systems under Severe Vibratory Stress: Application to Fatigue Testing of Military Vehicles” aims to advance the methodologies of the NF X 50-144 standard, a key framework for environmental testing in mechanical engineering. The research will harness artificial intelligence (AI) to improve vibration analyses, offering new insights into the resilience of complex systems under extreme conditions, with a focus on defense applications.

Vibration analysis plays a vital role in ensuring the durability and reliability of military vehicles and their embedded systems, such as armament, communication, and sensor technologies. These systems face intense and variable vibrational stresses from diverse operational environments, including rugged terrains and dynamic scenarios, which current testing approaches, based on simplified assumptions, often fail to fully capture. This project will explore AI-driven solutions to address these challenges, enhancing the ability to predict fatigue and mechanical performance through advanced techniques like classification of non-stationary signals and innovative modeling and classification approaches.

The research will tap into the potential of AI tools, such as autoencoders and other learning frameworks, to process complex vibration data and uncover patterns which traditional methods overlook. By refining how we assess and interpret these signals, the project aims to deliver more robust and adaptable solutions for evaluating system behavior. As AI becomes a cornerstone of future defense technologies, this work will position you at the forefront of innovation, contributing to both scientific progress and strategic advancements in military resilience.

Offered in collaboration with Inria, the University of Angers, and DGA Land Techniques, this PhD provides a unique platform for candidates passionate about AI, mechanical engineering, and defense. The outcomes will directly enhance the reliability of military vehicles, with broader implications for industries like aerospace and automotive, potentially shaping standards used by leading organizations. Based between Rennes and Angers, the project combines access to cutting-edge resources with a multidisciplinary team, offering an exceptional opportunity to make a meaningful impact in a high-stakes field.

Profil du candidat :
MSc (or soon-to-be) graduates

Formation et compétences requises :
This PhD, blending AI and mechanical resilience for military vehicles, seeks motivated MSc (or soon-to-be) graduates with the following qualifications:

Graduation Topics: Ideal profiles are those with backgrounds in computational mechanics, AI/data science and/or advanced scientific computing. Candidates can come from various MSc-level curriculums involving signal processing, machine learning, computational mechanics (vibration, fatigue, or structural dynamics), vehicle engineering (e.g., mechatronics), physics or applied mathematics. Experience in implementing numerical methods in high-level programming languages (Matlab, Python, Julia, …) is essential.
Academic Excellence: Outstanding curriculum with top-class grades, reflecting a strong academic track record. Candidates are expected to possess outstanding problem-solving abilities and a proven aptitude for teamwork.
Computer Literacy: Proficiency in navigating advanced algorithms and theoretical concepts, with strong analytical skills.
Citizenship: Should be an EU citizen due to the defense-related nature of the project.
Passion and Drive: Enthusiasm for AI, defense applications, and advancing engineering solutions.

Adresse d’emploi :
Centre Inria de l’Université de Rennes
Campus universitaire de Beaulieu
Avenue du Général Leclerc
35042 Rennes Cedex

Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : LITIS
Durée : 36 mois
Contact : Sebastien.Adam@univ-rouen.fr
Date limite de publication : 2025-04-30

Contexte :
Nous proposons une thèse financée au LITIS de l’INSA de Rouen.

Sujet :
cf PJ

Profil du candidat :
cf PJ

Formation et compétences requises :
cf PJ

Adresse d’emploi :
Laboratoire LITIS, INSA de rouen
76800 Saint Etienne du Rouvray

Document attaché : 202504091057_TheseDGNN.pdf

Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : LITIS
Durée : 36 mois
Contact : Sebastien.Adam@univ-rouen.fr
Date limite de publication : 2025-04-30

Contexte :
Nous proposons une thèse financée au laboratoire LITIS sur le sujet décrit dans la pièce jointe.

Sujet :
Cf. PJ

Profil du candidat :
Cf PJ

Formation et compétences requises :
cf PJ

Adresse d’emploi :
Laboratoire LITIS, UFR ST, Site du Madrillet
76800 Saint Etienne du Rouvray

Document attaché : 202504091055_TheseGapHIX.pdf