Conservatoire national des arts et métiers

About: Conservatoire national des arts et métiers is a education organization based out in Paris, France. It is known for research contribution in the topics: Population & Orthogonal frequency-division multiplexing. The organization has 3573 authors who have published 7127 publications receiving 141430 citations. The organization is also known as: CNAM & Conservatoire des arts et métiers.

Studying User Satisfaction with the MOOC Platform Interfaces Using the Example of Coursera and Open Education Platforms

Abstract: The advantages of online courses over traditional education lie in the availability of individual training programs, but now the problem of involving and keeping students on the course is urgent for MOOC. Thus, despite the fact that test results on the reasons for unsuccessful completion of online courses by students are different, one of the identified factors was the interface design, while the problem of the influence of the interface on users is not adequately studied. In the present paper, a methodology for studying two different MOOC online platforms was developed, in order to identify factors that are more favorable for user convenience, as well as to identify the best design solutions. A comparative analysis of two MOOC platforms (Coursera and Open Education) made it possible to identify the best behavioral patterns and understand which details of the design of the online platform should be paid attention to and which should not. After carrying out a theoretical analysis and forming the methodology of work, as well as performing an experiment, recommendations were formulated for the development of the user interface of the Open Education platform.

Augmenting Physical Models with Deep Networks for Complex Dynamics Forecasting

Abstract: Forecasting complex dynamical phenomena in settings where only partial knowledge of their dynamics is available is a prevalent problem across various scientific fields. While purely data-driven approaches are arguably insufficient in this context, standard physical modeling based approaches tend to be over-simplistic, inducing non-negligible errors. In this work, we introduce the APHYNITY framework, a principled approach for augmenting incomplete physical dynamics described by differential equations with deep data-driven models. It consists in decomposing the dynamics into two components: a physical component accounting for the dynamics for which we have some prior knowledge, and a data-driven component accounting for errors of the physical model. The learning problem is carefully formulated such that the physical model explains as much of the data as possible, while the data-driven component only describes information that cannot be captured by the physical model, no more, no less. This not only provides the existence and uniqueness for this decomposition, but also ensures interpretability and benefits generalization. Experiments made on three important use cases, each representative of a different family of phenomena, i.e. reaction-diffusion equations, wave equations and the non-linear damped pendulum, show that APHYNITY can efficiently leverage approximate physical models to accurately forecast the evolution of the system and correctly identify relevant physical parameters.

First determination of the Planck constant using the LNE watt balance

Abstract: After separate developments of the different elements with continuous characterizations and improvements, the LNE watt balance has been assembled. This paper describes the system in detail and gives its first measurements of the Planck's constant h. The value determined in air is h = 6.626 068 8(20) × 10−34 Js which differs in relative terms by −0.05 × 10−7 from the h90 value and by −1.1 × 10−7 from that of the 2010 CODATA adjustment of h. The relative standard uncertainty associated is 3.1 × 10−7.

Filter bank based multi-mode multiple access scheme for wireless uplink

Abstract: This paper investigates the potential of filter bank (FB) processing in the context of uplink multi-user access. First, a FB based scheme, conceptually similar to the single carrier frequency division multiple access (SC-FDMA) developed by 3GPP for uplink in the Long Term Evolution of UMTS, is analyzed. Specifically, a method for synthesizing spectrally well-localized uplink waveforms with low peak-to-average power ratio, using FB based multicarrier (FBMC) modulation in combination with FB spreading, is introduced. Secondly, the superior frequency selectivity of the FB approach, when combined with frequency sampling -designed subchannel/subband processing, is found to enable flexible and bandwidth efficient multi-mode uplink transmission with relaxed constraints on inter-user timing synchronization. The proposed concept allows different mobile terminals to operate in the reverse link simultaneously in multicarrier, SC-FDMA, or conventional single carrier mode according to attributes such as the required transmission power.