Sylvain Marie

Sylvain Marie
Schneider Electric · IoT & Digital Transformation

About

24
Publications
3,622
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
59
Citations

Publications

Publications (24)
Conference Paper
Full-text available
Learning the structure of Bayesian networks from data is a NP-Hard problem that involves optimization over a super-exponential sized space. Still, in many real-life datasets a number of the arcs contained in the final structure correspond to strongly related pairs of variables and can be identified efficiently with information-theoretic metrics. In...
Presentation
Full-text available
`pytest` is undoubtedly the most popular test framework for python. Its fixture and parametrization mechanisms, as well as its detailed hook API and vibrant plugin ecosystem make it a must-know for any developer wishing to create quality software. Some key limitations in its engine and API, however, prevent users from truly unleashing their testing...
Conference Paper
Industrial end users face an increasing need to reduce the risk of unexpected failures and optimize their maintenance. This calls for both short-term analysis and long-term ageing anticipation. At Schneider Electric we tackle those two issues using both Machine Learning and First Principles models, to provide advanced services for electrical distri...
Conference Paper
Full-text available
Focused retrieval retrieves and ranks sub-parts of documents according to their estimated relevance to a query. Many approaches akin to Structured Document retrieval exploit documents structure to effectively retrieve logic elements (titles, sections, etc...). Other approaches like Passage Retrieval aim at retrieving arbitrary length text unit (pas...
Conference Paper
Full-text available
Data-driven automatic fault detection and diagnostics (AFDD) have gained a lot of research attention in recent years. Many existing solutions need to learn from the fault operation data to be able to diagnose the faults. However, these data are usually not available in buildings. In this study we present a data-driven AFDD solution for Air Handling...
Article
Full-text available
Data-driven automatic fault detection and diagnostics (AFDD) have gained a lot of research attention in recent years. Many existing solutions need to learn from the fault operation data to be able to diagnose the faults. However, these data are usually not available in buildings. In this study we present a data-driven AFDD solution for Air Handling...
Poster
Score-Based Bayesian Network Structure Learning In Presence Of (Quasi-)Deterministic Relations
Patent
An improved data processing method for analyzing the energy consumption of a site from measurement data comprising: selecting data time series; segmenting the time series into sections; projecting the sections; displaying Projections of the data sections, - selection of at least a first group of sections, - establishment of a numerical classificati...
Conference Paper
This work proposes a temporal and frequential metric learning framework for a time series nearest neighbor classification. For that, time series are embedded into a pairwise space where a combination function is learned based on a maximum margin optimization process. A wide range of experiments are conducted to evaluate the ability of the learned m...
Patent
The invention relates to a method for determining the structure of an electricity distribution network (10), comprising a supply station (14) comprising one or more power supply outlets (18 1, 18 2) (16 1, 16 2, 16 3), the method comprises the following steps: a) acquiring a first set of data relating to the electrical energy consumed by each consu...
Conference Paper
Full-text available
This work proposes a temporal and frequential metric learning framework for a time series nearest neighbor classification. For that, time series are embedded into a pairwise space where a combination function is learned based on a maximum margin optimization process. A wide range of experiments are conducted to evaluate the ability of the learned m...
Conference Paper
Full-text available
Time series are complex data objects, they may present noise, varying delays or involve several temporal granularities. To classify time series, promising solutions refer to the combination of multiple basic metrics to compare time series according to several characteristics. This work proposes a new framework to learn a combination of multiple met...
Conference Paper
Full-text available
We introduce a new software extension for the open-source data mining software RapidMiner. The extension allows analysts to use the Octave language to build new operators in RapidMiner processes. The extension relies on a configurable pool of GNU Octave interpreters to handle concurrent access. It integrates in RapidAnalytics to schedule processes...
Poster
Full-text available
This poster was made with the HOMES team at Schneider Electric and presented along with a demo of the research prototypes for energy efficiency in buildings. (Apologies for not remembering who was presenting with me, please let me know if your name should appear here)
Method
This RFC defines a specific discovery base driver for interacting with DPWS networked devices in the OSGi™ environment and exposing OSGi™ services as DPWS devices on local networks. It answers to the technical requirements enumerated in the OSGi™ RFP 86. The specification work has been carried out in ANSO ITEA European project (http://www.itea-off...
Presentation
Full-text available
Combining OSGi™ Technology and Web Services to realize the Plug-n-Play Dream in the Home Network
Method
A specific driver specification is required with the arrival of the Devices Profile for Web Services specification. It will define how to discover and control DPWS devices and services within an OSGi™ framework. The specification must be built upon the Discovery Base Driver concept defined in the OSGi™ Device Access specification.
Preprint
We present a semi-supervised algorithm to learn an interpolating function with minimal Laplacian semi-norm on a graph. This algorithm is directly inspired by the classical 'method of relaxations' used in physics to get discrete approximate solutions to the Dirichlet problem. We argue that this method is actually a gradient descent minimizing the en...
Conference Paper
Full-text available
L'exploitation des données collectées par les capteurs enfouis dans les entreprises offre de nouvelles opportunités économiques aux équipementiers. Cependant, la mise en place des services associés est actuellement encore faite de manière empirique. Le projet PISE1 s'est intéressé à proposer des méthodes et des outils pour la mise en place flexible...

Network

Cited By

Projects

Project (1)
Archived project
Collaboration between LIG AMA, GIPSA-Lab and Schneider Electric, on metric learning for timeseries.