Le département Science des Données rassemble quatre équipes d’expertises reconnues et complémentaires, couvrant la modélisation, la collection, la gestion, l’analyse et la construction de données et de connaissances (A&O, Bioinfo, LaHDAK, Rocs), permettant d’explorer les synergies entre expertises liées aux données, à l’apprentissage et à l’optimisation, notamment en lien avec les domaines de la bio-informatique, de l’IoT et des graphes de données.
Des traces numériques de l’ensemble des activités humaines sont aujourd’hui disponibles dans tous les domaines, des données souvent massives, hétérogènes, dynamiques et de qualité variable (les 4 V du Big Data : Volume, Variété, Vélocité, Véracité). Leur exploitation conduit à la définition d’un quatrième paradigme scientifique : la conception et la validation d’hypothèses, de modèles théorique et d’algorithmes, guidées par les données et en interaction avec les experts du domaine. Le département Science des Données s’intéresse à répondre de façon robuste aux défis des 4V, en termes de passage à l’échelle face au volume et à la vélocité des données, et de résistance aux biais de diversité et de qualité. Ces objectifs définissent de nouvelles problématiques informatiques dans le domaine du stockage, de la communication, de l’analyse et de l’optimisation des traitements, de l’interrogation et de l’enrichissement des données, de la découverte des connaissances et de l’apprentissage de modèles.
Avec près de 40 chercheurs et enseignants-chercheurs, le département Sciences des Données couvre un spectre large de thématiques fondamentales et en lien avec les applications : bases de données, fouille de données, Web sémantique, représentation des connaissances, algorithmique, combinatoire, optimisation stochastique et distribuée, apprentissage statistique et réseaux neuronaux, réseaux de communication, simulation. Il dispose également d’une expertise approfondie de la recherche interdisciplinaire et du dialogue avec les experts des domaines d’applications (particulièrement en biologie, médecine, sciences humaines et sociales, physique expérimentale), permettant un accès privilégié aux données d’intérêt et à l’évaluation des modèles et des algorithmes.
Coordination
Algorithmes, apprentissage et calcul, Sciences des Données
Jyotishka Das, Aaron Mamann, Francesca Bugiotti, Paul-Henry Cournède. Multimodal Survival Analysis on Hierarchical Representation of Clinical Trial Data. European Conference on Advances in Databases and Information Systems, ADBIS 2026, Sep 2026, Orleans, France, France. pp.117-126, ⟨10.1007/978-3-032-39823-9_11⟩. ⟨hal-05795334⟩
Théophile Sanchez, Pierre Jobic, Cyril Regan, Paul Verdu, Guillaume Charpiat, et al.. Attentive-SPIDNA: Attention-based neural networks for population genetics. 2026. ⟨hal-05688902⟩
Alexis Delplace, Xinyi Li, Samer Lahoud, Nur Zincir-Heywood. HalifaxDT: A Wireless Digital Twin from Open Geospatial Data. Workshop on Digital Twins over NextG Wireless Networks (DTwin 2026), Oct 2026, Austin, United States. pp.1-9, ⟨10.1145/3842434.3844958⟩. ⟨hal-05787661⟩
Nicolas Saint-Léger, Nicolas Férey, Joe Raad, Patrick Bourdot. Measuring the Degree of Multimodality: Toward a Foundation for Adaptive Extended Reality. 2026. ⟨hal-05779425⟩
Nicolas Saint-Léger, Nicolas Férey, Joe Raad, Patrick Bourdot. Sémantisation du contenu et du contexte pour la gestion de l’expérience utilisateur et de la multimodalité en environnement de Réalité Virtuelle et Augmentée. AFXR. Journées Françaises de la Réalité Étendue (jf·XR 2024), Oct 2024, Évry-Courcouronnes, France. 2024. ⟨hal-05779106⟩
Bruna Lopes, Gabriel Schwartz, Sylvain Chevallier, Raphael Y de Camargo, Bruno Aristimunha. Learning aligned EEG representations with subject-specific encoders. Scientific Reports, 2026, ⟨10.1038/s41598-026-69220-z⟩. ⟨hal-05758650⟩
Hugo De Oliveira, Lucas Foissey, Yousef Shnaiwer, Megumi Kaneko. Energy-Aware Pruning for Federated Deep Reinforcement Learning in Multi-Interface IoT Networks. IEEE Internet of Things Journal, 2026, 13 (4), pp.6713-6726. ⟨10.1109/JIOT.2025.3639564⟩. ⟨hal-05753287⟩
Thémis Lemarchand, Fanny Pouyet, Maud Fagny. Detecting polygenic selection signatures using tree sequences statistics. Journée des doctorants du département BAP, May 2026, Versailles, France. ⟨hal-05757195⟩
Léo Planche, Anna Ilina, María C Ávila-Arcos, Flora Jay, Emilia Huerta-Sanchez, et al.. An archaic reference-free method to jointly infer Neanderthal and Denisovan introgressed segments in modern human genomes. Molecular Biology and Evolution, 2025, ⟨10.1093/molbev/msag235⟩. ⟨hal-05747173⟩
Marco Rosario Capodiferro, Léo Planche, Emily M Breslin, Linda Ongaro, María C Ávila-Arcos, et al.. Archaic ancestry inference in imputed ancient human genomes. Nature Communications, 2026, 17 (1), pp.9480. ⟨10.1038/s41467-026-76204-0⟩. ⟨hal-05747161⟩
Dylan Sechet, Marc Evrard, Matthieu Kowalski. Simulation-Based Inference for Plate Reverb System Identification. DAFX 2026 – 29th International Conference on Digital Audio Effects, Sep 2026, Cambridge, United States. ⟨hal-05742718⟩
Brahim Driss, Alex Davey, Riad Akrour. PB$^2$: Preference Space Exploration via Population-Based Methods in Preference-Based Reinforcement Learning. RLC 2026 – Third Reinforcement Learning Conference, Aug 2026, Montreal, Canada. ⟨hal-05732524⟩
Yuchen Wang, Ghali Laraqui Houssaini, Yuchen Hou, Shwetha Salimath, Sylvain Wlodarczyk, et al.. Accelerating Industrial Geological Stratification from Well Logs with Deep Learning and Statistical Constraints. BigData – 2025 IEEE International Conference on Big Data, Deep Learning, Transformers, Seq2Seq, Statistical Distribution, Oil & Gas Extraction, Dec 2025, Macau, China. pp.2663-2668, ⟨10.1109/BigData66926.2025.11402336⟩. ⟨hal-05720044⟩
Bruno Aristimunha, Dung Truong, Pierre Guetschel, Seyed Yahya Shirazi, Isabelle Guyon, et al.. EEG Foundation Challenge: From Cross-Task to Cross-Subject EEG Decoding. NeurIPS 2025 – 39th Annual Conference on Neural Information Processing Systems, Dec 2025, San Diego, United States. ⟨hal-05714360⟩
Bruno Aristimunha, Ce Ju, Antoine Collas, Florent Bouchard, Ammar Mian, et al.. SPD Learn: A Geometric Deep Learning Python Library for Neural Decoding Through Trivialization. 2026. ⟨hal-05714371⟩
Beranger Ovono-Ekore, J. Deparis, Moncef Hidane, Julien Mille, Farid Smaï, et al.. Physics-Consistent Neural Inversion of 2-D DC Resistivity Data. NSG 2026: 32nd Meeting of Environmental and Engineering Geophysics, European Association of Geoscientists and Engineers (EAGE), Sep 2026, Thessaloniki, Greece. pp.1-5, ⟨10.3997/2214-4609.202620086⟩. ⟨hal-05706633⟩
Lilia Izri, Benoît Groz, Silviu Maniu. Implementing Efficient Linear Bandits Via Sketches and Random Projections. BigData – 2025 IEEE International Conference on Big Data, Dec 2025, Macau, China. pp.956-965, ⟨10.1109/BigData66926.2025.11401692⟩. ⟨hal-05684798⟩
Alan Gany, Bogdan Cautis, Silviu Maniu. Structural Adversarial Attacks on Relational Deep Learning under Integrity Constraints. AIDB 2026 – VLDB Workshop on Databases and AIArtificial Intelligence, Aug 2026, Boston, United States. ⟨hal-05687375⟩
Jonathan Colin, Silviu Maniu. Maximizing Diverse Information Exposure in Content-Based Social Networks. Transactions on Large-Scale Data- and Knowledge-Centered Systems, 2026, Lecture Notes in Computer Science, 16240, pp.1-31. ⟨10.1007/978-3-662-72449-1_1⟩. ⟨hal-05687331⟩
Benedictus Kent Rachmat, Thomas Gerald, Zheng Zhang, Cyril Grouin. QA Analysis in Medical and Legal Domains: A Survey of Data Augmentation in Low-Resource Settings. ACL 2025 – 63rd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop), ACL, Jul 2025, Vienna, Austria. pp.1132-1144, ⟨10.18653/v1/2025.acl-srw.89⟩. ⟨hal-05683004⟩
Flavio Tuteri, Sergio Chibbaro, Alexandros Alexakis. Intermittency in Shell Models of Turbulent Cascades: from Single-Branch to Multi-Branch. 2026. ⟨hal-05679006⟩
Bruno Aristimunha, Igor Carrara, Pierre Guetschel, Sara Sedlar, Pedro Rodrigues, et al.. Mother of all BCI Benchmarks. 2026, ⟨swh:1:dir:6d89a182d4e69572a73784c6616ef1cee5493d77⟩. ⟨hal-05675918⟩
Bruno Aristimunha Pinto. Learning Representations of Electroencephalogram using Deep Learning. Machine Learning [cs.LG]. Université Paris-Saclay; Universidade Federal do ABC, 2026. English. ⟨NNT : 2026UPASG016⟩. ⟨tel-05668261⟩
Marie Schmit, Melvin Selim Atay, Khalid Belhajjame, Ulysse Le Clanche, Emmanuel Coquery, et al.. ShareFAIR-KG, a centralised knowledge base of scientific workflows. JOBIM 2026 – Journées Ouvertes en Biologie, Informatique et Mathématiques, Jun 2026, Strasbourg, France. ⟨hal-05666980⟩
Eva Boguslawski. Politiques d’apprentissage par renforcement pour l’exploitation de réseaux électriques avec actions continues dans des architectures centralisées et décentralisées. Intelligence artificielle [cs.AIArtificial Intelligence]. Université Paris-Saclay, 2026. Français. ⟨NNT : 2026UPASG028⟩. ⟨tel-05657134⟩
Hugo Thimonier, José Lucas de Melo Costa, Fabrice Popineau, Arpad Rimmel, Bich-Liên Doan. T-JEPA: Augmentation-Free Self-Supervised Learning for Tabular Data. ICLR 2025 – International Conference on Learning Representations, Apr 2025, Sinagpore, Singapore. ⟨10.48550/arXiv.2410.05016⟩. ⟨hal-05624276⟩
Sarah Cohen-Boulakia, Frédéric Lemoine, George Marchment, Marine Djaffardjy, Alban Gaignard, et al.. Reuse and reproduce bioinformatic pipelines using scientific workflow systems. Berlin Universities Publishing. Workflow systems for large-scale scientific data analysis, Berlin Universities Publishing, 2026, 978-3-98781-067-1. ⟨10.14279/depositonce-25828⟩. ⟨pasteur-05640482⟩
Jean-Baptiste Malagnoux, Matthieu Kowalski. Connecting the Dots Between CDL and NMF: A Shared Framework for Time-Frequency Decomposition. IEEE Transactions on Signal Processing, 2026, 74, pp.2448-2460. ⟨10.1109/TSP.2026.3698440⟩. ⟨hal-05635856⟩
Fabrice Popineau. Repenser l’évaluation à l’heure de l’iaIntelligence Artificielle générative : changer son objet plutôt que son format. 2026. ⟨hal-05630830⟩
Hanane Kteich, Gianluca Quercini, Joe Raad, Fatiha Saïs. BEAM: A First Benchmark for Microdata Entity Alignment with Knowledge Graphs. The 41st ACM/SIGAPP Symposium On Applied Computing, Mar 2026, Thessaloniki, Greece. ⟨10.1145/3748522.3779966⟩. ⟨hal-05609694⟩
Lucas José Velôso de Souza, Ingrid Valverde Reis Zreik, Adrien Salem-Sermanet, Nacéra Seghouani, Lionel Pourchier. A Deep Learning-Based Approach for Mangrove Monitoring. ECML PKDD 2024 – Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Sep 2024, Vilnius, Lithuania. pp.242-253, ⟨10.1007/978-3-032-25311-8_20⟩. ⟨hal-05622381⟩
Victor Spitzer. Data-driven stochastic optimization for green hydrogen production planning and related single-item lot-sizing problems. Operations Research [math.OC]. Université Paris-Saclay, 2026. English. ⟨NNT : 2026UPASG022⟩. ⟨tel-05620127⟩
Lucas Heck dos Santos, João Guilherme Prado Barbon, Sylvain Chevallier, Denis G Fantinato. Non-Linear Activation Functions for Deep Riemannian Neural Networks. ESANN 2026 – 34th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, Apr 2026, Bruges, Belgium. pp.535-540, ⟨10.14428/esann/2026.ES2026-312⟩. ⟨hal-05616889⟩
Li-Qing Jiang, Amanda Fay, Jens Daniel Müller, Luke Gregor, Alizée Roobaert, et al.. Synthesis of data products for ocean carbonate chemistry. Earth System Science Data, 2026, 18 (2), pp.1405 – 1462. ⟨10.5194/essd-18-1405-2026⟩. ⟨hal-05607765⟩
Ali Kadhum Idrees, Tara Ali-Yahiya, Sara Kadhum Idrees, Raphael Couturier. DaTOS: Data Transmission Optimization Scheme in Tactile Internet-based Fog Computing Applications. International Symposium on Personal, Indoor and Mobile Radio Communications, Sep 2022, Kyoto, Japan. pp.01-06, ⟨10.1109/PIMRC54779.2022.9977593⟩. ⟨hal-04257370⟩
Simon Coumes, Pierre-Henri Paris, François Schwarzentruber, Fabian Suchanek. Qiana: A First-Order Formalism to Quantify over Contexts and Formulas with Temporality. Journal of Artificial Intelligence Research, 2026, 85, pp.12:1-12:35. ⟨10.1613/jair.1.18402⟩. ⟨hal-05575513⟩
Arnaud Quelin, Frederic Austerlitz, Flora Jay. Using supervised machine learning methods to infer demographic history from genomic data. Machine Learning for Evolutionary Genomics Data, May 2024, Heraklion, Greece. ⟨hal-05570974⟩
Arnaud Quelin, Jazeps Medina Tretmanis, Emilia Huerta-Sanchez, Frederic Austerlitz, Flora Jay. Assessing the contribution of ancient genomic data to the inference of historical demographic parameters. JOBIM 2025 – Journées Ouvertes en Biologie, Informatique et Mathématiques, Jul 2025, Bordeaux, France. ⟨hal-05570966⟩
Jean Cury, Théophile Sanchez, Erik Madison Bray, Jazeps Medina-Tretmanis, María Ávila-Arcos, et al.. Inferring effective population sizes of bacterial populations while accounting for unknown recombination and selection: a deep learning approach. ECML PKDD 2022 – Machine Learning for Microbial Genomics workshop, Sep 2022, Grenoble, France. ⟨hal-05554282⟩
George Marchment, Sarah Cohen-Boulakia, Frédéric Lemoine. Computational Reproducibility With Scientific Workflows: Analysing viral genomes with Nextflow. REP’25 ACM Conference on Reproducibility and Replicability, Jul 2025, Vancouver, Canada. ⟨hal-05525260⟩
Cyriaque Rousselot, Olivier Allais, Philippe Caillou, Julia Mink, Florian Yger. Pesticide Aerial Concentration : Estimations Toward Population Health Impact Assessment. E4H – Engineering for Health Annual Forum 2025, Nov 2025, Palaiseau (Ecole Polytechnique), France. ⟨hal-05447129⟩
Enrico Marchesini, Eva Boguslawski, Alessandro Leite, Christopher Amato, Matthieu Dussartre, et al.. MARL2GRID-TR: A multi-agent RL benchmark in power grid operations. ICLR 2026 – 14th International Conference on Learning Representations, Apr 2026, Rio de Janeiro, Brazil. ⟨hal-05532479⟩
Jonathan Colin. Enabling Diversity of Content Sources in Online Social Networks. Social and Information Networks [cs.SI]. Université Paris-Saclay, 2025. English. ⟨NNT : 2025UPASG100⟩. ⟨tel-05523131⟩
Marie Schmit, Ulysse Le Clanche, George Marchment, Sarah Cohen-Boulakia, Olivier Dameron, et al.. ShareFAIR-KG. 2025, ⟨swh:1:rev:a96c28345857df13aafbc85461df2ab0cf259ef8;origin=https://gitlab.liris.cnrs.fr/sharefair/knowledge_base_workflow_annotations/ShareFAIR-KG.git;visit=swh:1:snp:e53f57ef45e971e1b9b76477d1f39f4fbe3de77f⟩. ⟨hal-05517690⟩
Louis Ollivier, Gilles Fischer, Fanny Pouyet. A Robust Computational Framework to Characterize the Genetic Diversity Across 3,570 Strains in S. cerevisiae. JOBIM 2025 – Journées Ouvertes en Biologie, Informatique et Mathématiques, Jul 2025, Bordeaux, France. ⟨hal-05484804⟩
Louis Ollivier, Gilles Fischer, Fanny Pouyet. A Robust Computational Framework to Characterize the Genetic Diversity Across 3,570 Strains in Saccharomyces cerevisiae. Yeast 2025, Jul 2025, Paris, France. ⟨hal-05484807⟩
Ludovic Arnold, Sébastien Rebecchi, Sylvain Chevallier, Hélène Paugam-Moisy. An Introduction to Deep Learning. European Symposium on Artificial Neural Networks (ESANN), Apr 2011, Bruges, Belgium. ⟨hal-01352061⟩
Jean-Baptiste Malagnoux. Apprentissage de dictionnaires convolutifs et factorisation non négative de matrices pour la séparation de sources et les problèmes inverses. Mathématiques [math]. Université Paris-Saclay, 2025. Français. ⟨NNT : 2025UPASG096⟩. ⟨tel-05472451v2⟩
Fatiha Saïs, Caio Corro, Gaël Lejeune, Dominique Longin. PFIA 2025. Bulletin de l’Association Française pour l’Intelligence Artificielle, 130, 2025, Association Française pour l’Intelligence Artificielle. ⟨hal-05480308⟩
Antoine Szatkownik. Modélisation générative dans un espace latent et évaluation de données synthétiques en génomique des populations. Informatique [cs]. Université Paris-Saclay, 2025. Français. ⟨NNT : 2025UPASG109⟩. ⟨tel-05475569v2⟩
Burak Yelmen, Merve Nur Güler, Tõnu Kollo, Märt Möls, Guillaume Charpiat, et al.. Bias in genome-wide association test statistics due to omitted interactions. RECOMB 2026 – 30th Annual International Conference on Research in Computational Molecular Biology, May 2026, Thessaloniki, Greece. ⟨10.1101/2025.11.21.689603⟩. ⟨hal-05474710⟩
Michael Tangermann, Sylvain Chevallier, Matthias Dold, Pierre Guetschel, Reinmar Kobler, et al.. Learning from small datasets — review of workshop 6 of the 10th International BCI Meeting 2023. Journal of Neural Engineering, 2025, 22 (3), pp.033001. ⟨10.1088/1741-2552/addf80⟩. ⟨hal-05462858⟩
Giovanni Catania, Aurélien Decelle, Cyril Furtlehner, Beatriz Seoane. A theoretical framework for overfitting in energy-based modeling. ICML 2025 – 42nd International Conference on Machine Learning, Jul 2025, Vancouver, Canada. pp.6891-6919. ⟨hal-05466288⟩
El-Hacen Diallo, Khaldoun Al Agha, Steven Martin. TRADE-5G: A blockchain-based transparent and secure resource exchange for 5G network slicing. Blockchain: Research and Applications, 2025, 6 (1), pp.100246. ⟨10.1016/j.bcra.2024.100246⟩. ⟨hal-05460607⟩
Khaldoun Al Agha, Guy Pujolle. Intelligent Energy Arbitration in Edge Networks with Distributed Digital Twins. WD’25 – 2025 13th Wireless Days Conference, Dec 2025, Niterói, Rio de Janeiro, Brazil. pp.1-7, ⟨10.1109/WD67713.2025.11302632⟩. ⟨hal-05460627⟩
Abdulhakeem Abdulazeez, Nicola Zema, Tara Ali-Yahiya, Steven Martin. Edge-Enhanced Attention Mechanism for Dynamic UAV Formation Fault Detection. ISNCC – 2025 International Symposium on Networks, Computers and Communications, Oct 2025, Paris, France. pp.1-8, ⟨10.1109/ISNCC66965.2025.11250475⟩. ⟨hal-05458459⟩
Panayiotis Panayiotou, Audrey Poinsot, Alessandro Leite, Nicolas Chesneau, Marc Schoenauer, et al.. CausalProfiler: Generating Synthetic Benchmarks for Rigorous and Transparent Evaluation of Causal Machine Learning. 2026. ⟨hal-05455135⟩
Khaldoun Al Agha, Pauline Loygue, Guy Pujolle. Preparing for the 6G Era: Introducing the Internet of Edges. NOMS 2024-2024 IEEE Network Operations and Management Symposium, May 2024, Seoul, South Korea. pp.1-3, ⟨10.1109/NOMS59830.2024.10575595⟩. ⟨hal-05450988⟩
Kinda Khawam, Hussein Taleb, Samer Lahoud, Hassan Fawaz, Dominique Quadri, et al.. Non-cooperative edge server selection game for federated learning in IoT. Network Operations and Management Symposium (NOMS), May 2024, Seoul, South Korea. pp.1-6, ⟨10.1109/NOMS59830.2024.10575406⟩. ⟨hal-05450907⟩
Yeabsira Asefa Ashengo, Tara Ali Yahiya, Nicola Roberto Zema. Efficient Task Offloading in Multi-access Edge Computing Servers using Asynchronous Meta Reinforcement Learning in 5G. 2024 IEEE Symposium on Computers and Communications (ISCC), Jun 2024, Paris, France. pp.1-6, ⟨10.1109/ISCC61673.2024.10733620⟩. ⟨hal-05451050⟩
El-Hacen Diallo, Mohameden Dieye, Omar Dib, Pierre Valiorgue. An agnostic and secure interoperability protocol for seamless asset movement. Journal of Network and Computer Applications (JNCA), 2024, 230, pp.103930. ⟨10.1016/j.jnca.2024.103930⟩. ⟨hal-05451144⟩