é du LIP6 dernières (Jean-Luc Mounier) (Aurore Marcos)Tue, 16 Jan 2018 00:00:00 GMTWed, 10 Jan 2018 12:46:48 GMT360frTextWrangler, TexMate, code maison PHP, MySQLCopyright 11/2004, 02/2006, 01/2008 Laboratoire d'Informatique de Paris 6 Félicitations à Jean-Gabriel Ganascia Colloquium-Schwartz Colloquium : Jean-Luc Schwartz (CNRS), mardi 16 janvier à 18h. Info-143 Jean-Gabriel Ganascia, l'IA et moi sur France Culture. Info-139 Eleni Diamanti de l’équipe QI vient d’obtenir un financement européen prestigieux (ERC Starting). Info-142 Eleni Diamanti de l’équipe QI sur France Culture Colloquium-Roscoe Colloquium : Timothy Roscoe (ETH Zürich), jeudi 30 novembre à 18h. Organise-937 Circadian Mood Variations in Twitter Content, jeudi 18 janvier à 10h30. Circadian regulation of sleep, cognition, and metabolic state is driven by a central clock, which is in turn entrained by environmental signals. Understanding the circadian regulation of mood, which is vital for coping with day-to-day needs, requires large datasets and has classically utilised subjective reporting. We use a massive dataset of over 800 million Twitter messages collected over the course of 4 years in the United Kingdom. We extract robust signals of the changes that happened during the course of the day in the collective expression of emotions and fatigue. We use methods of statistical analysis and Fourier analysis to identify periodic structures, extrema, change-points, and compare the stability of these events across seasons and weekends. We reveal strong, but different, circadian patterns for positive and negative moods. The cycles of fatigue and anger appear remarkably stable across seasons and weekend/weekday boundaries. Positive mood and sadness interact more in response to these changing conditions. Anger and, to a lower extent, fatigue show a pattern that inversely mirrors the known circadian variation of plasma cortisol concentrations. Most quantities show a strong inflexion in the morning. Since circadian rhythm and sleep disorders have been reported across the whole spectrum of mood disorders, we suggest that analysis of social media could provide a valuable resource to the understanding of mental disorder.<br/><br/>**Bio**<br/><br/>Fabon defended his PhD thesis in Computer Science in 2013 “on Learning and Representation from Texts for both Emotional and dynamical Information” at the University of Pierre et Marie Curie, in the DAPA department at LIP6. After graduating he held a short post-doctoral position in LIP6, working on building interpretable models for the classification of multivariate time series’ data. At this time he grew an interest in the analysis of time series’ data, and in the Fourier transform as a mean to extract meaningful features from data. He later joined the University of Bristol as a research associate in 2014 where he worked on efficient machine learning algorithms for data streams, and developed tools to study our human behaviours at a collective level via the analysis of the social media and large samples of press archives. He combined his works on information dynamics and his interest in the study of emotions to research periodic patterns of emotions and mental health. His results provide evidence that a share of the variance in our collective behaviours and emotions are predictable across the year, and over the 24-h cycle.<br/><br/>_Plus d'information sur Fabon Dzogang : <>_<br/> (benjamin piwowarski) Séminaire Donnees et APprentissage Artificiel Mon, 08 Jan 2018 18:23:39 GMT Organise-938 CPUs and GPUs: can we get the best of both worlds?, jeudi 18 janvier à 10h30. GPUs are now established parallel accelerators for high-performance <br/>computing applications and machine learning. Part of their success is <br/>based on their so-called SIMT execution model. SIMT binds together <br/>threads of parallel applications so they perform the same instruction at <br/>the same time, in order to execute their instructions on <br/>energy-efficient SIMD units. Unfortunately, current GPU architectures <br/>lack the flexibility to work with standard instruction sets like x86 or <br/>ARM. Their implementation of SIMT requires special instruction sets with <br/>control-flow reconvergence annotations, and they seldom support complex <br/>control flow like exceptions, context switches and thread migration.<br/><br/>In this talk, I will present how we can generalize the SIMT execution <br/>model of GPUs to general-purpose processors and multi-thread <br/>applications, and then use it to design new CPU-GPU hybrid cores. These <br/>hybrid cores will form the building blocks of heterogeneous <br/>architectures, that will combine CPU-like cores and GPU-like cores which <br/>all share the same instruction set and programming model. Beside <br/>improving programmability, generalized SIMT enables key improvements <br/>that were not possible in the traditional SIMD model, such as <br/>simultaneous execution of divergent paths or guarantees of fairness. It <br/>opens the way for a whole spectrum of new architectures, hybrids of <br/>latency-oriented superscalar processors and throughput-oriented SIMT GPUs.<br/><br/> (Marc Mezzarobba) GT Pequan Wed, 10 Jan 2018 12:46:48 GMT Organise-902 21th ICIN Conference Innovations in Clouds, Internet and Networks, mardi 20 février. ICIN 2018 will be held 20-22 February 2018 at Orange Gardens premises near Paris, France. The ICIN conference, established in 1989 and praised for the richness of its social networking, is also recognized for its continuous adaptation to emerging technical trends.<br/><br/>2018 will be the year of coordinated 5G deployment across all EU Member states, targeting early network introduction, and moving towards commercial large scale introduction by the end of 2020 at the latest.<br/><br/>As the main theme for ICIN 2018, “5G Network Architecture and Solutions: are we ready?” will explore the maturity and gaps remaining for end-to-end and large scale 5G technologies deployment. The conference will foster presentation of novel approaches and technical solutions, as well as experimental validation and proof of concepts. In addition to this edition theme, ICIN 2018 will keep a wide range of topics of interest within the fields of Cloud, Internet, and Networks as listed below.<br/><br/>Topics of interest:<br/><br/>5G Control and Management Architecture<br/>XaaS Architectures and Models<br/>Network solutions for IoT and Vertical Industries<br/>CDN and ICN Control and Architecture<br/>Autonomic, Cognitive and Intent-based Networking and Management<br/>Network Artificial Intelligence, Machine Learning and Reasoning<br/>Software Defined Networking<br/>Network Function Virtualization<br/>Fog, Edge and Multi-Access Computing<br/>Multi-domain, Federated and Hierarchical Orchestration<br/>5G Functional Decomposition and Infrastructure Slicing<br/>Service Function Chaining<br/>Virtual Network Embedding and Function Placement<br/>Traffic Engineering and Performance Management for ultra-low latency<br/>Advanced real-time communications<br/>Performance, QoS/QoE and Scalability Analysis<br/>Resource Dimensioning and Optimization<br/>Security, Trust and Privacy<br/>Big and Fast Data Management<br/>Measurement, Monitoring and Telemetry<br/>Protocols for Programmable Networks and Software-Defined Infrastructure<br/>Open source and Open standards (guy pujolle) ICIN 2018 Fri, 02 Jun 2017 13:06:54 GMT