BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//LORIA - ECPv6.17.2//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:LORIA
X-ORIGINAL-URL:https://www.loria.fr
X-WR-CALDESC:Évènements pour LORIA
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:20200329T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20201025T010000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20210328T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20211031T010000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20220327T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20221030T010000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Europe/Paris:20210308T100000
DTEND;TZID=Europe/Paris:20210308T113000
DTSTAMP:20210304T195532Z
CREATED:20210304T195503Z
LAST-MODIFIED:20210304T195532Z
UID:11667-1615197600-1615203000@www.loria.fr
SUMMARY:D3 seminar: Rémy Grünblatt and Mourad Dridi
DESCRIPTION:Next Department 3 seminar will take place on Monday\, 8th March at 10am with two presentations. \n10h00: De l’évaluation des performances Wi-Fi à la mobilité contrôlée pour les réseaux de drones\, par Rémy Grünblatt (INRIA Inria Lille – Nord Europe) \nFrom Wi-Fi performance evaluation to controlled mobility for drone networks \nMobility in telecommunication networks is often seen as a hassle that needs to be dealt with: a mobile wireless device has to adapt is trans-mission parameters in order to remain connected to its counterpart(s)\,as the channel evolves with the device’s movements. Drones are no exception. In this presentation\, I’ll show how controlled mobility can be used to increase the performance of a drone network\, with a focus on IEEE 802.11 networks and small multi-rotor drones.\nI will first describe how mobility is partially dealt with in 802.11 networks\, that is to say using rate adaptation mechanisms\, and I’ll present the rate adaptation algorithm used in the Intel Wi-Fi chipsets\, also used in the Intel Aero Drone\, which I reverse engineered. The study of this rate adaptation algorithm\, thanks to its implementation in the network simulator NS-3\, allows its comparison against other well-known algorithms\, and highlights how big the impact of such algorithms is\, for drone networks.\nI will then present a controlled mobility solution which leverages the antenna radiation pattern of the drones. This algorithm is evaluated thanks to a customized simulation framework for antenna and drone simulation\, based on NS-3. This solution\, which works with any rate adaptation algorithm\, is distributed\, and do not require a global coordination that would be costly. It also does not require a full and complete control of the drone mobility as existing controlled mobility solutions require\, which makes this solution compatible with various applications. \n10h40: Robustesse des architectures pluri-cœurs et NoCs\, par Mourad Dridi (Université Polytechnique Hauts-de-France\, INSA\, Valenciennes) \n\nMon projet de recherche va porter principalement sur l’applicabilité de la théorie de l’ordonnancement\ntemps réel dans la conception et l’implémentation des systèmes embarqués.\nLes applications d’aujourd’hui demandent un fonctionnement en temps-réel\, pour fournir des résultats\ndans des délais bien précis\, comme la navigation par GPS\, le codage du son dans un téléphone cellulaire\, la\ncommande du gouvernail d’un avion\, la vidéo surveillance active\, l’ABS d’une voiture\, …\n\nMon objectif est donc de développer des méthodes\, techniques et outils pour la conception\, la vérification et la réalisation\nde systèmes informatiques temps réel. Afin d’atteindre cet objectif\, je voudrais exercer des activités de recherche dans les sujets suivants :\n\n\nLes systèmes temps réel multi-critiques et les systèmes temps réel à criticité mixte\nLe problème du placement dans les architectures pluri-cœurs\nL’ordonnancement des communications dans les architectures pluri-cœurs (NoCs\, SpaceWire …)
URL:https://www.loria.fr/event/d3-seminar/
CATEGORIES:Séminaire
END:VEVENT
BEGIN:VEVENT
DTSTART;VALUE=DATE:20210323
DTEND;VALUE=DATE:20210327
DTSTAMP:20210312T122217Z
CREATED:20210312T122217Z
LAST-MODIFIED:20210312T122217Z
UID:11730-1616457600-1616803199@www.loria.fr
SUMMARY:Journées nationales de l'informatique Mathématique du GDR-IM
DESCRIPTION:Les Journées Nationales de l’Informatique Mathématique (JNIM) sont organisées chaque année par le GDR IM (Informatique Mathématique). C’est une occasion d’information et d’échange annuelle entre les chercheurs du GDR et de la discipline. Elles auront lieu du 23 au 26 mars. \nInscriptions gratuites et obligatoires avant le 18 mars. \nProgramme complet sur le site de la conférence.
URL:https://www.loria.fr/event/journees-nationales-de-linformatique-mathematique-du-gdr-im/
CATEGORIES:Séminaire
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Paris:20210325T110000
DTEND;TZID=Europe/Paris:20210325T120000
DTSTAMP:20210323T143513Z
CREATED:20210323T143331Z
LAST-MODIFIED:20210323T143513Z
UID:11820-1616670000-1616673600@www.loria.fr
SUMMARY:MALOTEC Séminaire : Explainability for Sequential Decision-Making
DESCRIPTION:Machine learning has been used to aid decision-making in several domains\, from healthcare to finance. Understanding the decision process of ML models is paramount in high-stakes decisions that impact people’s lives\, otherwise\, loss of control and lack of trust may arise. Often\, these decisions have a sequential nature. For instance\, the transaction history of a credit card must be considered when predicting the risk of fraud of the most recent transaction Although RNNs are state-of-the-art models for many sequential decision-making tasks\, they are perceived as black-boxes\, creating a tension between accuracy and interpretability. While there has been considerable research effort towards developing explanation methods for ML\, recurrent models have received relatively much less attention. Recently\, Lundberg and Lee unified several methods under a single family of additive feature attribution explainers. From this family\, KernelSHAP has seen a wide adoption throughout the literature; however\, this explainer is unfit to explain models in a sequential setting\, as it only accounts for the current input not the whole sequence. In this work\, we present TimeSHAP\, a model-agnostic recurrent explainer that builds upon KernelSHAP and extends it to sequences. TimeSHAP explains recurrent models by computing feature-\, timestep-\, and cell-level attributions\, producing explanations at both the feature and time axes. As sequences may be arbitrarily long\, we further propose two pruning methods that are shown to dramatically decrease TimeSHAP’s computational cost and increase its reliability. We validate TimeSHAP by using it to explain predictions of two RNN models in two real-world fraud detection tasks\, obtaining relevant insights into these models and their predictions. \n\n\nSpeaker:\nJoão Bento\, Research Data Scientist at Feedzai working on explainability for Machine Learning models.\n\nOn TEAMS. More information at:\nhttps://malotec.loria.fr
URL:https://www.loria.fr/event/malotec-seminaire-explainability-for-sequential-decision-making/
LOCATION:online
CATEGORIES:Séminaire
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Paris:20210325T160000
DTEND;TZID=Europe/Paris:20210325T170000
DTSTAMP:20210316T154038Z
CREATED:20210304T173118Z
LAST-MODIFIED:20210316T154038Z
UID:11664-1616688000-1616691600@www.loria.fr
SUMMARY:DigiTrust Webinar: Alfredo Cuzzocrea
DESCRIPTION:The first webinar of LUE IMPACT project DigiTrust will take place on Thursday\, 25th March at 4pm. \nProfessor Alfredo Cuzzocrea\, Excellence Chair in Computer engineering financed by DigiTrust\, will give a presentation entitled “Privacy-Preserving Big Data Management and Analytics in Distributed Environments: Models\, Issues\, Proposals” \nThe webinar will take place on Teams. \nAbstract: \nNowadays\, big data management and analytics is gaining momentum within the research community. Basically\, the main issue with big data management concerns with effectively and efficiently managing massive big data repositories for a wide variety of typical data management tasks\, such as representation\, querying\, indexing\, partitioning\, and so forth. On the other hand\, big data analytics concerns with extracting useful\, actionable knowledge from big data repositories for decision making purposes\, by extending classical approaches inherited from decades of data mining and machine learning research. In this so-delineated context\, the issue of supporting privacy-preserving big data management and analytics plays a first-class role\, especially with respect to the wide class of emerging big data application scenarios\, which range from social networks to bio-informatics\, from sensors networks to web recommendation tools\, from e-science systems to e-government systems\, and so forth. In all these applicative settings\, protecting the privacy of sensitive information\, for instance personal data or aggregate data\, can be clearly intended as an enabling technology. Distributed environments are the natural humus for collocating privacy-preserving big data management and analytics tasks\, including the astonishing raise of blockchain technology. Among others\, Cloud systems play the major role\, even stirred-up by recent technological advancements that have really enhanced the ICT industry at now. More and more today\, real-life Cloud-based applications\, such as smart cities\, intelligent transportation systems\, marketplace tools and so forth\, are indeed posing new challenges to privacy-preserving big data research\, thus contributing to improve the scientific area. \nThis seminar will explore the research challenge represented by supporting privacy-preserving big data management and analytics in distributed environments\, by exploring models and issues\, and describing some proposed solutions. In particular\, the seminar will consider the special case of supporting privacy-preserving OLAP analytics in distributed environments. This kind of big data analytics tools predicates the definition of multidimensional metaphors to be embedded into the analytics phase\, and supports the achievement of richer actionable knowledge (prone to decision making) to be extracted from distributed big data repositories. On the other hand\, the issue of supporting privacy preservation within OLAP analytics is still an open research problem\, whit many advancements still to be achieved. According to this consideration\, an innovative privacy-preserving OLAP analytics in distributed environments approach is presented in this seminar\, along with the proposal of some interesting extensions that are at now under evolution.
URL:https://www.loria.fr/event/digitrust-webinar/
CATEGORIES:Séminaire
END:VEVENT
END:VCALENDAR