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

Laboratoire/Entreprise : LISTIC – Université Savoie-Mont-Blanc
Durée : 18 mois
Contact : jean-yves.ramel@univ-smb.fr
Date limite de publication : 2026-08-29

Contexte :
Graph Neural Networks for Smart Monitoring of Bio-based Composites
(GRINCOMP MIAI Project)

Sujet :
The project aims to develop advanced AI methods based on Graph Neural Networks (GNNs) for the monitoring and predictive maintenance of bio-based composite structures.

More info : https://miai-cluster.univ-grenoble-alpes.fr/job-opportunities/postdoctoral-positions/

Profil du candidat :
Your profile
• PhD in Machine Learning, Data Science, Applied Mathematics, or related fields
• Strong experience in deep learning (experience with GNNs is a plus)
• Solid programming skills in Python (PyTorch, TensorFlow, PyG, etc.)
• Interest in interdisciplinary research and real-world applications

Formation et compétences requises :
Nice to have:
• Experience with sensor data, IoT, or physics-informed ML
• Background or interest in materials science or structural monitoring
• Time-series analysis / signal processing

Adresse d’emploi :
Location: Campus Technolac – Chambery – France
• Start date: End of 2026

Research environment
You will work in a collaborative and interdisciplinary setting involving:
• LISTIC laboratory (AI, Machine Learning , Graphs, data science)
• SYMME laboratory (materials science, smart composites)
• Interactions with national academic and industrial partners

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

Laboratoire/Entreprise : LITIS-INSA Rouen
Durée : 3 ans
Contact : aomar.osmani@insa-rouen.fr
Date limite de publication : 2026-09-01

Contexte :
Thèse financée dans le cadre des allocations de recherche état/région.

Sujet :
Meta-Learning and Artificial General Intelligence for a
Computational Theory of Assistance to Human Learning

Profil du candidat :
Nous recherchons un(e) candidat(e) issu(e) d’un M2 ou diplôme d’ingénieur en informatique, data science, IA ou sciences cognitives computationnelles, en mathématiques avec une forte appétence pour
la recherche.

Compétences souhaitées :
— bases solides en ML/DL ;
— intérêt pour les sciences cognitives, les sciences de l’éducation, ou l’optimisation ;
— goût pour la modélisation mathématique et pour la modélisation et la programmation ;

— des connaissances en méta-apprentissage, RL, modèles séquentiels (RNN/Transformers) consti-
tuent un plus.

Environnement :
— Projet pluridisciplinaire (IA, sciences cognitives, ingénierie pédagogique) à fort impact sociétal ;
— ressources de calcul et données pour des expérimentations à grande échelle ;
— valorisation attendue dans des conférences internationales (NeurIPS, ICLR, AIED, etc.).

Formation et compétences requises :
ML/DL, programmation (Python), expérience PyTorch/TensorFlow appréciée ;

intérêt pour éducation/cognition ; méta-learning/RL/modèles séquentiels

Adresse d’emploi :
INSA de Rouen
685 Avenue de l’Université 76800 Saint-Etienne-du-Rouvray

Document attaché : 202602171414_sujetAnglais(1).pdf

Offre en lien avec l’Action/le Réseau : SaD-2HN/– — –

Laboratoire/Entreprise : Laboratoire d’Informatique de Grenoble
Durée : 3
Contact : camille.bernard@univ-grenoble-alpes.fr
Date limite de publication : 2026-12-30

Contexte :
Human societies rely on geographic information to describe physical (e.g., schools, roads, cycle lanes) or virtual (e.g., administrative zoning) features that constitute their territories. In Geographic Information Science, the representation of such geographic entities is well established. However, representing and contextualizing the changes these entities undergo over time—driven, for example, by urban policies promoting low-carbon mobility—remains a major challenge.
These complex phenomena, also referred to as territorial transformations or spatial-temporal changes, involve one or several geographic entities that evolve at a given time or over a defined period under various pressures. Today, documenting these transformations in a machine-readable and computationally exploitable form is still largely absent.
In urban analysis, geographic databases describe—through standards such as CityGML or other formats—the built environment, transport networks, and urban infrastructure. Some open and collaborative databases such as OpenStreetMap (OSM) are often more up to date than official datasets due to continuous user contributions, thereby indirectly capturing urban evolution. Another more marginal initiative, OpenHistoricalMap (OHM), uses the OSM framework to build an open, editable historical map. However, none of these databases or standards explicitly model or document the changes undergone by geographic entities (e.g., transport networks, streets, cycle lanes, etc.).
Although some datasets, such as the IGN BD TOPO, provide differences between successive versions, these differences do not qualify or semantically describe the nature of the changes. Consequently, monitoring urban evolution remains difficult due to the lack of standards for representing territorial transformations. Yet, adopting a shared vocabulary for describing urban changes would make it possible to quantify them, define indicators of urban evolution, and facilitate comparisons between cities with similar or divergent trajectories.
In response to the lack of structured data describing territorial changes, and the absence of standardized tools for representing such transformations in urban planning and land-use monitoring, the objective of this PhD is to automate the detection and description of territorial changes, and to build temporal sequences of these changes in order to characterize territorial trajectories. These change catalogues will support interoperability between systems and provide a foundation for broader community adoption. They will take the form of Spatiotemporal Knowledge Graphs (ST-KGs), designed according to FAIR principles (Findable, Accessible, Interoperable, Reusable), ensuring accessibility and reuse, notably for digital twins of territories and predictive AI systems aimed at forecasting urban evolution based on historical data.

Sujet :
Geographic databases such as OpenStreetMap (OSM) and the IGN BD TOPO are continuously evolving to reflect territorial transformations. Identifying, describing, and characterizing the changes occurring in these geospatial databases over time is a major challenge for understanding and analyzing the evolution of territories.
This PhD is part of the ANR GEvoK project (Geographic Entities Evolution in Knowledge Graphs), which aims at the automatic detection and semantic annotation of geographic changes from heterogeneous data sources, including satellite imagery and collaborative datasets such as OpenStreetMap.
The proposed approach combines artificial intelligence techniques for the automatic detection of temporal changes, Semantic Web technologies for representing and enriching the knowledge derived from these detections, and large language models (LLMs) for querying the resulting knowledge bases and generating natural language narratives describing observed changes.

Profil du candidat :
The candidate must hold a Master’s degree in Computer Science, Geomatics, Data Science, or Artificial Intelligence.
•

Formation et compétences requises :

• Strong knowledge of Python and machine learning (classification, anomaly detection, clustering);
• Knowledge of Semantic Web technologies and knowledge representation (RDF, OWL, SPARQL);
• Basic knowledge of geographic data processing; familiarity with QGIS and spatial data formats (GeoJSON, shapefile, etc.);
• Autonomy, rigor, analytical skills, and interest in hybrid AI / geospatial approaches.
Required level of French: Upper intermediate (B2)
Required level of English: Upper intermediate (B2)

Adresse d’emploi :
LIG, Bâtiment IMAG, 700 Av. Centrale, 38401 Saint-Martin-d’Hères

Document attaché : 202607031244_offre-these-gevok-26-en.pdf

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

Laboratoire/Entreprise : ISIR – Institut des Systèmes Intelligents et de Ro
Durée : 36 mois
Contact : patrick.gallinari@sorbonne-universite.fr
Date limite de publication : 2026-12-30

Contexte :
AI4Science is an emerging research field that investigates the potential of AI methods to advance scientific discovery, particularly through the modeling of complex natural phenomena. This fast-growing area holds the promise of transforming how research is conducted across a broad range of scientific domains. One especially promising application is in modeling complex dynamical systems that arise in fields such as climate science, earth science, biology, and fluid dynamics. A diversity of approaches is currently being developed, but this remains an emerging field with numerous open research challenges in both machine learning and domain-specific modeling.

This PhD project aims to investigate the next generation of AI models for physical dynamics. The objective is to develop generative world models that learn structured representations of physical systems and can efficiently model, predict, and reason about their evolution. The research will focus on applications such as fluid mechanics and climate science while addressing fundamental questions at the intersection of machine learning and scientific computing.

Sujet :
The main objective of this PhD is to develop generative world models for physical dynamics that combine scalability, uncertainty modeling, and scientific consistency.

The research will explore several complementary directions:

Learning transferable representations of physical dynamics, by developing latent representations that capture the underlying structure of physical systems and can generalize across multiple physical regimes and downstream tasks.

Generative modeling of physical trajectories, using recent approaches such as diffusion models, flow matching, and stochastic interpolants to represent uncertainty, multimodality, and long-term evolution of complex dynamical systems.

Physically consistent generative models, by integrating physical constraints and scientific priors into generative learning in order to produce solutions that remain both accurate and scientifically valid.
The exact research direction will be adapted to the candidate’s interests and background and may emphasize either methodological developments or applications to scientific domains such as fluid dynamics and climate modeling.

Profil du candidat :
Computer science or applied mathematics. Good programming skills.

Formation et compétences requises :
Master degree in computer science or applied mathematics, Engineering school. Background and experience in machine learning.

Adresse d’emploi :
Sorbonne Université (S.U.), Pierre et Marie Campus in the center of Paris. The candidate will integrate the MLIA team (Machine Learning and Deep Learning for Information Access) at ISIR (Institut des Systèmes Intelligents et de Robotique).

Document attaché : 202607061524_2026-05-PhD-Description-Generative-World-models-Physics.pdf