BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Etis - ECPv6.16.2//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Etis
X-ORIGINAL-URL:https://www.etis-lab.fr
X-WR-CALDESC:Events for Etis
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:Europe/Paris
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20230326T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20231029T010000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20240331T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20241027T010000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20250330T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20251026T010000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20260329T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20261025T010000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Europe/Paris:20250325T143000
DTEND;TZID=Europe/Paris:20250325T153000
DTSTAMP:20260707T191029
CREATED:20250311T162315Z
LAST-MODIFIED:20250311T162315Z
UID:9046-1742913000-1742916600@www.etis-lab.fr
SUMMARY:Data&AI Seminar : Ozgun Pinarer
DESCRIPTION:Titre : Analyse et Prévision de Données Météorologiques Distribuées Utilisant l’Apprentissage Fédéré et TinyML\nRésumé :\nCe séminaire présentera une étude approfondie sur l’application de l’apprentissage fédéré aux systèmes embarqués à ressources limitées dans les stations météorologiques. L’objectif principal est d’évaluer la performance des modèles d’apprentissage local et fédéré en tenant compte de critères tels que la précision\, la consommation d’énergie et l’utilisation de la mémoire. Notre approche intègre des techniques d’apprentissage automatique et profond pour traiter des données météorologiques collectées à partir de stations en Corse. Ce séminaire abordera la méthodologie adoptée\, les défis liés à l’utilisation de l’apprentissage fédéré sur des systèmes embarqués et les perspectives d’amélioration dans le domaine de l’IoT pour l’analyse météorologique. \nBiographie :\nOzgun Pinarer est maître de conférences et directeur adjoint du département de génie informatique de l’Université Galatasaray à Istanbul\, en Turquie. Il a obtenu son diplôme en génie informatique à l’Université Galatasaray en 2010\, puis a poursuivi ses études en obtenant une maîtrise en génie informatique à la même institution en 2012. En 2017\, il a obtenu son doctorat à l’INSA Lyon\, au laboratoire LIRIS. Ses recherches portent sur l’Internet des Objets (IoT)\, le calcul embarqué optimisé pour le matériel\, la gestion des flux de données des capteurs dans les environnements intelligents\, l’optimisation énergétique des systèmes IoT\, la maintenance prédictive et la gestion des actifs\, ainsi que sur le Tiny ML et l’apprentissage fédéré pour les applications IoT. Il a mené plusieurs projets académiques et industriels dans ces domaines\, notamment sur la gestion des données IoT\, la communication en cas de catastrophe via LoRa\, ainsi que sur l’apprentissage automatique et fédéré appliqué aux systèmes embarqués à faible consommation énergétique. Il organise également\, depuis 2018\, une session spéciale sur les données de santé lors de la conférence IEEE Big Data. Ses travaux récents portent sur l’application de modèles d’apprentissage fédéré aux capteurs météorologiques embarqués\, avec pour objectif d’améliorer la précision des prévisions tout en optimisant les ressources des dispositifs à faible consommation.
URL:https://www.etis-lab.fr/event/dataai-seminar-ozgun-pinarer/
LOCATION:CYU Saint-Martin\, salle de réunion A551\, 2 avenue Adolphe-Chauvin\, Cergy Pontoise\, France
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.etis-lab.fr/wp-content/uploads/2025/03/ozgun-pinarer.jpg
ORGANIZER;CN="Hajer Baazaoui":MAILTO:hajer.baazaoui@ensea.fr
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Paris:20250306T140000
DTEND;TZID=Europe/Paris:20250306T160000
DTSTAMP:20260707T191029
CREATED:20250304T155300Z
LAST-MODIFIED:20250304T155300Z
UID:9025-1741269600-1741276800@www.etis-lab.fr
SUMMARY:IA multimodale et IA générative pour l’indexation vidéo pédagogique
DESCRIPTION:Boris Borzic\, ingénieur de recherche CNRS à ETIS (équipe Data&AI et pôle ingénierie) interviendra à la MSH Mondes de Nanterre sur le thème “IA multimodale et IA générative pour l’indexation vidéo pédagogique“. \nRésumé : \nDans un contexte d’apprentissage adaptatif\, la stratégie « vidéo first » promet une expérience pédagogique plus fluide. Cependant\, la gestion efficace de ces contenus nécessite des outils innovants pour faciliter la navigation et l’accès aux informations clés. L’application d’intelligence artificielle (IA) multimodale et générative peut répondre à ce besoin en permettant une indexation vidéo pédagogique plus efficace. Les questions de révision corrélées à des réponses sous forme d’extrait vidéo\, accompagnées de techniques de RAG (Retrieval Augmented Generation\, génération augmentée de récupération )\, offrent un potentiel significatif pour améliorer l’expérience de l’apprenant et la personnalisation de l’apprentissage. De plus le montage streaming en temps réel sans exportation/importation vidéo constitue une stratégie efficace pour optimiser les ressources et réduire les coûts de bande passante. \nBoris Borzic\, PhD Research engineer\, Ingénieur de recherche CNRS et Fondateur de la startup Deeptech Sequencia (labellisé CNRS RISE)\nETIS – Equipes Traitement de l’Information et Systèmes UMR 8051 / ENSEA – CNRS – CY Cergy Paris Université \nBibliographie :\n \n\nTharsan Senthivel\, Ngoc-Son Vu\, Boris Borzic. Detection Transformer with Diversified Object Queries. IEEE ICIP 2023\, Oct 2023\, Kuala Lampur\, Malaysia. ⟨hal-04304226⟩\nMarianne Froye\, Olivier Belin\, Julien Longhi\, Boris Borzic\, Claudia Marinica\, et al.. L’IDHN : une structure innovante au service de la polysémie du numérique. Humanistica 2020\, May 2020\, Bordeaux\, France. ⟨hal-02875614⟩\nClaudia Marinica\, Julien Longhi\, Nader Hassine\, Abdulhafiz Alkhouli\, Boris Borzic. #Idéo2017 : une plateforme citoyenne dédiée à l’analyse des tweets lors des événements politiques. Extraction et Gestion des Connaissances (EGC)\, Jan 2018\, Paris\, France. ⟨hal-01699423⟩\nDalia Saigh\, Boris Borzic\, Abdulhafiz Alkhouli\, Julien Longhi. A Linguistic Contribution to an Automatic Classification of Communities and their Analysis. Questions de communication\, 2017\, 31\, pp.161 – 182. ⟨10.4000/questionsdecommunication.11097⟩. ⟨hal-01793225⟩\nAbdulhafiz Alkhouli\, Dan Vodislav\, Boris Borzic. Continuous Top-k Queries in Social Networks. CoopIS 2016\, 2016\, Rhodes\, Greece. pp.24 – 42\, ⟨10.1007/978-3-319-48472-3_2⟩. ⟨hal-01417787⟩\n\nInformations pratiques : \nHeure: 6 mars 2025 02:00 PM Paris\nMSH Mondes\, Université Paris Nanterre (Bâtiment Max Weber\, salle de séminaire 1)\nhttps://cnrs.zoom.us/j/91688466917?pwd=8Snj6106A4uV2ALvprhLjefUWI5arz.1\nID de réunion: 916 8846 6917 / Code secret: x5Yk2t \n 
URL:https://www.etis-lab.fr/event/ia-multimodale-et-ia-generative-pour-lindexation-video-pedagogique/
LOCATION:MSH Monde\, 21 allee de l'université bâtiment maw weber\, nanterre\, 92000\, France
CATEGORIES:Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Paris:20250303T110000
DTEND;TZID=Europe/Paris:20250303T123000
DTSTAMP:20260707T191029
CREATED:20250217T084116Z
LAST-MODIFIED:20250217T084116Z
UID:9014-1740999600-1741005000@www.etis-lab.fr
SUMMARY:Seminar ETIS-ICI: Rodrigo C. de Lamare
DESCRIPTION:Rodrigo C. de Lamare\, from PUC-RIO\, will give an invited talk on Monday\, March 3rd\, 2025\, 11:00 am\, ENSEA\, room 384. Please find below the details. \nZoom : https://cnrs.zoom.us/j/94161301799?pwd=bTgHYnHGmIqeG3BHPyJEM429aXzVSy.1 \nTitle: Energy-efficient distributed and federated learning for IoT networks \nAbstract:\nIn this presentation\, we will present an energy-efficient distributed learning framework using coarsely quantized signals for Internet of Things (IoT) networks. In particular\, we develop distributed quantization-aware least-mean\, recursive least-squares and federated learning algorithms that can learn parameters in an energy-efficient fashion using signals quantized with few bits while requiring a low computational cost. Moreover\, we develop a bias compensation strategy to further improve the performance of the proposed learning algorithms. We carry out a statistical analysis of the proposed algorithms and derive analytical expressions for predicting the mean-square deviation. A computational complexity evaluation and a study of the power consumption of the proposed and existing techniques are then presented. Numerical results assess the proposed learning algorithms against existing techniques for parameter estimation tasks in IoT networks. \nBiography:\nRodrigo C. de Lamare was born in Rio de Janeiro\, Brazil\, in 1975. He received his Diploma in electronic engineering from the Federal University of Rio de Janeiro in 1998 and the MSc and PhD degrees in electrical engineering from the Pontifical Catholic University of Rio de Janeiro (PUC-Rio) in 2001 and 2004\, respectively. Since January 2006\, he has been with the Communications Research Group\, Department of Electronic Engineering\, University of York\, United Kingdom\, where he is a Professor. Since April 2013\, he has also been a Professor at PUC-RIO. Dr de Lamare is a senior member of the IEEE. He has served as editor for IEEE Wireless Communications Letters\, IEEE Signal Processing Letters and IEEE Transactions on Communications and currently serves as associate editor of IEEE Transactions on Signal Processing. His research interests lie in communications and signal processing\, areas in which he has published over 550 papers in international journals and conferences.
URL:https://www.etis-lab.fr/event/seminar-etis-ici-rodrigo-c-de-lamare/
LOCATION:ENSEA\, salle 384\, 6 avenue du Ponceau\, Cergy\, 95000\, France
CATEGORIES:Seminar
ORGANIZER;CN="Sara Berri":MAILTO:sara.berri@ensea.fr
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Paris:20250220T144500
DTEND;TZID=Europe/Paris:20250220T163000
DTSTAMP:20260707T191029
CREATED:20250210T100905Z
LAST-MODIFIED:20250210T100905Z
UID:8995-1740062700-1740069000@www.etis-lab.fr
SUMMARY:ETIS Tandem Talk: Paul Gay & Guillaume Renton
DESCRIPTION:Tandem Talk by Paul Gay (UPPA) and Guillaume Renton (ETIS) on Sustainability and Machine Learning. \nPaul Gay | Interactions between sustainibility and machine learning research\nAI is a controversial topic and different visions of sustainibility co-exist in the IT communities. In this talk\, I will describe the state of the art from the GreenIT community to assess environmental impact of IT projects. Although the main ideas\, such as life cycle analysis\, description of embodied impacts and indirect effects have been explored for 10-20 years\, we are only beginning to see them applied to AI systems and projects. \nIn a second part\, I will present two of my current machine learning applications to sustainibility topics. The first one is the use of active learning to detect and classify controversial topic on renewable energies in social networks. The second one is to exploit the technique of early exit\, an interesting tool where the amount of compute depends on the data\, and which find applications in edge-cloud settings. \nGuillaume Renton | Reducing computation costs without jeopardizing precision of Entity Alignment tasks\nIn recent work\, we took interest in the computational cost of one of our entity alignment model\, HybEA. The idea was to provide an efficiency analysis on top of a performance analysis\, which is the main source of comparison between AI models. The efficiency analysis was conducted by using fvcore in order to compute the number of GFLOPS required to train the model. This has led to surprising results\, showing that the initial embedding sizes of the models were oversized. This allowed us to greatly reduce the computational cost of our model with a very small loss of accuracy. \n  \nThe talks will take place in the Curium at ENSEA as well as online:\nhttps://cnrs.zoom.us/j/92031778182?pwd=BcsbEGTQf2JwG8jIUkpaK35fZ1WXUK.1
URL:https://www.etis-lab.fr/event/etis-tandem-talk-paul-gay-guillaume-renton/
LOCATION:ENSEA\, Curium\, avenue du Ponceau\, Cergy\, 95014\, France
CATEGORIES:ETIS,Seminar
ORGANIZER;CN="Camille Simon Chane":MAILTO:camille.simon-chane@ensea.fr
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Paris:20250207T100000
DTEND;TZID=Europe/Paris:20250207T113000
DTSTAMP:20260707T191029
CREATED:20250124T072146Z
LAST-MODIFIED:20250124T072146Z
UID:8972-1738922400-1738927800@www.etis-lab.fr
SUMMARY:Training neural models using logic: results\, challenges\, and applications
DESCRIPTION:DATA&AI Team Seminar (Online) \nEfi Tsamoura\, Senior Researcher at Samsung AI \nTitle: Training neural models using logic: results\, challenges\, and applications\nAbstract: Neurosymbolic learning (NSL) vows to transform AI by combining the strong induction capabilities of neural models with the strong deduction capabilities of symbolic knowledge representation and reasoning techniques. This talk centers around an NSL problem that has received significant attention lately: training neural classifiers using supervision produced by logical theories. Empirical research has shown the advantages of this learning setting over end-to-end deep neural architectures in multiple aspects\, including accuracy and model complexity. Despite the extensive empirical research\, limited theoretical analysis has been dedicated to understanding if and under which conditions we can learn the underlying neural models. \nThis talk covers this gap by proposing necessary and sufficient conditions\, which ensure that we can learn the underlying models under rigorous guarantees. I will also discuss the relationship between this problem and other known problems in the machine learning literature. Furthermore\, I will present new challenges inherent to this NSL setting and propose solutions to overcome those challenges\, leading to models with substantially higher accuracy. I will conclude this talk with recent applied results and open challenges. \n  \nBio: Efi Tsamoura is a Senior Researcher at Samsung AI\, Cambridge\, UK. In 2016\, she was awarded a prestigious early career fellowship from the Alan Turing Institute\, UK\, for her work on logic and databases\, and before that\, she was a Postdoctoral Researcher in the Department of Computer Science of the University of Oxford. Her main research interests lie in the areas of logic\, knowledge representation and reasoning\, and neurosymbolic learning\, while her recent outcomes involve scaling symbolic reasoning to billions of triples\, as well as addressing open problems in neuro-symbolic learning. Her research has been published in top-tier machine learning\, AI\, and database venues (NeurIPS\, ICML\, SIGMOD\, VLDB\, PODS\, AAAI\, IJCAI\, etc.).
URL:https://www.etis-lab.fr/event/training-neural-models-using-logic-results-challenges-and-applications/
LOCATION:Zoom
CATEGORIES:Seminar
ORGANIZER;CN="Vassiis Christophides":MAILTO:vassilis.christophides@ensea.fr
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Paris:20241015T133000
DTEND;TZID=Europe/Paris:20241015T150000
DTSTAMP:20260707T191029
CREATED:20241001T142628Z
LAST-MODIFIED:20241015T084144Z
UID:8568-1728999000-1729004400@www.etis-lab.fr
SUMMARY:Séminaire DATAAI - Minh Ha Quang
DESCRIPTION:An optimal transport and information geometric framework for Gaussian processes\nAbstract:\nInformation geometry (IG) and Optimal transport (OT) have been attracting much research attention in various fields\, in particular machine learning and statistics. In this talk\, we present results on the generalization of IG and OT distances for finite-dimensional Gaussian measures to the setting of infinite-dimensional Gaussian measures and Gaussian processes. Our focus is on the Entropic Regularization of the 2-Wasserstein distance and the generalization of the Fisher-Rao distance and related quantities. In both settings\, regularization leads to many desirable theoretical properties\, including in particular dimension-independent convergence and sample complexity. The mathematical formulation involves the interplay of IG and OT with Gaussian processes and the methodology of reproducing kernel Hilbert spaces (RKHS). All of the presented formulations admit closed form expressions that can be efficiently computed and applied practically. The mathematical formulations will be illustrated with numerical experiments on Gaussian processes. \nBio:\nMinh Ha Quang is the team leader of the Functional Analytic Learning team in the RIKEN Center for Advanced Intelligence Project (AIP)\, Tokyo\, JAPAN. He received his PhD in mathematics from Brown University (Providence\, RI\, USA) under the supervision of Stephen Smale. Before joining RIKEN\, he was a researcher at the Pattern Analysis and Computer Vision group at the Italian Institute of Technology (Istituto Italiano di Tecnologia) in Genoa (Genova)\, Italy. Prior to Italy\, he was a postdoctoral researcher at the University of Vienna\, Austria\, and the Humboldt University of Berlin\, Germany. His current research interests focus on machine learning and statistical methodologies using theories and techniques from Functional Analysis and related mathematical fields. In particular\, he has been working on theories and methods involving reproducing kernel Hilbert spaces (RKHS)\, Riemannian geometry\, Matrix and Operator Theory\, Information Geometry\, and Optimal Transport\, especially in the Infinite-Dimensional setting. \nTeams link: https://teams.microsoft.com/l/meetup-join/19%3ameeting_YjA3NjcxODItZTdjNy00Yzg0LThkYjQtNTg2MDhlYmEwMjY0%40thread.v2/0?context=%7b%22Tid%22%3a%22aa8bdaa4-8feb-46c0-b5e6-31c96337579b%22%2c%22Oid%22%3a%22c4e82db7-e9d5-4310-b5c4-91f9189f0cba%22%7d \n 
URL:https://www.etis-lab.fr/event/seminaire-dataai-minh-ha-quang/
LOCATION:ENSEA\, salle 331\, 6 avenue du Ponceau\, Cergy\, 95000\, France
CATEGORIES:ETIS,Seminar
ORGANIZER;CN="Xuan-Son Nguyen":MAILTO:xuan-son.nguyen@ensea.fr
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Paris:20241010T153000
DTEND;TZID=Europe/Paris:20241010T170000
DTSTAMP:20260707T191029
CREATED:20241001T074009Z
LAST-MODIFIED:20241004T165217Z
UID:8566-1728574200-1728579600@www.etis-lab.fr
SUMMARY:Présentation "tandem" ETIS
DESCRIPTION:Sur le thème des “low-techs”\, présentation tandem\, avec Aurélien Béranger (UTC)\, et Arnaud Blanchard (ETIS). \nAurélien Béranger est doctorant en sciences de l’information et de la communication\, le titre provisoire de sa thèse est “Écologies du faire. Les discours et la matière de la hiérarchisation technologique dans le mouvement low-tech”. \nPrésentation de Aurélien Béranger\nMouvement low-tech\, communautés innovantes\, et outils numériques d’appropriation technologique \nRésumé : L’objet de cette présentation est de donner un aperçu de recherches doctorales visant à investiguer ce qui se cache sous la formule « low-tech ». Elle sera structurée en deux temps. Tout d’abord\, le concept de low-tech sera resitué dans la généalogie de la promotion de technologies alternatives et le mouvement low-tech\, tel qu’il se développe en France depuis une quinzaine d’années\, sera présenté dans toute sa diversité. Dans un second temps\, seront présentées des cas d’études de communautés de pratiques développant des technologies dites low-tech. Il s’agira\, outre de donner un aperçu sur les formes communautaires d’innovation dans le domaine matériel\, de présenter les places qui sont faites aux outils numériques dans ces espaces et comment ces derniers peuvent être développés pour répondre à des enjeux d’appropriation et de démocratisation des technologies. \n 
URL:https://www.etis-lab.fr/event/presentation-tandem-etis/
LOCATION:ENSEA\, Curium\, avenue du Ponceau\, Cergy\, 95014\, France
CATEGORIES:ETIS,Seminar
ORGANIZER;CN="Camille Simon Chane":MAILTO:camille.simon-chane@ensea.fr
END:VEVENT
END:VCALENDAR