Papers by Juliette MATTIOLI
AI and Ethics, Jan 7, 2024
HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific r... more HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. Copyright
HAL (Le Centre pour la Communication Scientifique Directe), Mar 10, 2023
This booklet gathers 52 papers, either in the form of articles or posters, presented during the s... more This booklet gathers 52 papers, either in the form of articles or posters, presented during the second edition of the Confiance.ai Days held in Saclay on October 4-6, 2022. Altogether they give a good snapshot of the research and development work done in the Confiance.ai program, an industrial and academic initiative of the national Grand Challenge on provable and certifiable AI, launched in support of the France 2030 strategy. Among these papers, a dozen are presented as « external contributions » that were selected by an ad-hoc committee following a call for papers. All other communications belonged to one of five so-called « villages » with physical implementation in the conference hall, distributing the work done in Confiance.ai into five topics : « End-to-end approach » ; « from Operational Design Domain to Data » ; « Explainability and Understanding» ; « Robustness and Monitoring » ; « Embedded AI ». After two years of activity, and complementing the Confiance.ai white paper, this document shows the diversity and the quality of the work done in the programme. Some important and up-to-date subjects are addressed, such as -only to name a few -out-of-distribution detection, adversarial robustness, semi-or self-supervised learning, explainability by design, verification and validation, embedded AI etc. We hope that you will enjoy reading parts of this document as much as we enjoyed preparing and attending the 2022 Confiance AI days.
Springer eBooks, 2001
In this paper, we present a domain specific optimization framework based on a concurrent model-ba... more In this paper, we present a domain specific optimization framework based on a concurrent model-based approach for handling the complete problem of mapping a DSP application on a parallel architecture. The implementation is based on Constraint Programming and the model is described in details. Our concurrent resolution approach undertaking linear and non linear constraints takes advantage of the special features of signal processing applications. Finally, our mapping tool developped with the Eclair solver is evaluated and compared to a classical approach.
Organizations need to face the rapid growth of social networks and its adoption by their collabor... more Organizations need to face the rapid growth of social networks and its adoption by their collaborators. While organizations require collaborative tools to get their work done, their collaborators seem to prefer social tools like Facebook or Twitter. We present in this article a new kind of organizational Content Management System that takes care of both documentary and social sources of knowledge.
Proceedings of SPIE, Aug 1, 1991
ABSTRACT The purpose of this paper is to show that some adapted preprocessing can help neural net... more ABSTRACT The purpose of this paper is to show that some adapted preprocessing can help neural networks in classifying binary planar shapes. After recalling some basic notions of mathematical morphology and some shape numbers, a morphological preprocessing transforming a binary planar shape into a curve which is translation, rotation, and scale invariant, namely, the spectral function, is presented. These extracted shape features are discriminated by statistical and neural methods. Finally, an application on real data is described.
With Lean 4.0, Symbolic AI (Artificial Intelligence) and data-driven AI are becoming a cornerston... more With Lean 4.0, Symbolic AI (Artificial Intelligence) and data-driven AI are becoming a cornerstone to Total production Maintenance (TMP). We present in this paper is a brief overview of various AI techniques which are use to improve TMP in order to address some preventive, predictive and prescriptive maintenance issues.
In a System Engineering perspective, asset management (AM) is related to a subset of techniques f... more In a System Engineering perspective, asset management (AM) is related to a subset of techniques focusing on the in-service phase, aligned with product life-cycle management discipline. Today, within AM solution market, the integration of Artificial Intelligence (AI) technics above traditional entreprise solution is a key trend. This paper is focusing on how symbolic AI and data driven AI could improve some issues of the AM life cycle, in particular in asset acquisition, performance analysis and forecasting, asset monitoring, predictive and prescriptive maintenance, supply chain optimisation including spare parts management…
Computers in Human Behavior, 2014
The wide adoption of social and connected tools in organizations leads them to think again about ... more The wide adoption of social and connected tools in organizations leads them to think again about their behavior regarding how they manage their resources. They now consider the resources users can produce on various social media and how correctly index them in the organization knowledge base. We present in this paper the model of a digital ecosystem, which permits the indexing of either documentary resources or those produced on a social platform with the help of an ontology.
HAL (Le Centre pour la Communication Scientifique Directe), Jul 1, 2021
Ai Magazine, Jun 13, 2023
HAL (Le Centre pour la Communication Scientifique Directe), Jan 31, 2012
2013 Eighth International Conference on P2P, Parallel, Grid, Cloud and Internet Computing, Oct 1, 2013
With the popularisation of Web 2.0 applications, organizations see in them a means to improve the... more With the popularisation of Web 2.0 applications, organizations see in them a means to improve their productivity by giving new and efficient collaborative tools to their members. The juxtaposition of several different Web applications inside organisation Information System (IS) can be viewed as a digital ecosystem. We show in this article how the members' activities inside these digital ecosystems can help to compute the social network of the organization. To do so, we present a semantic model of such an ecosystem and a method to compute the social network of the organization from the different resources the users share on the different Web applications deployed in the IS of the organization. We will finally present an implementation of our model and our algorithm in order to compute the social network of the french Knowledge Engineering (KE) community.
Lean Manufacturing is well known as an effective means to improve productivity and decrease costs... more Lean Manufacturing is well known as an effective means to improve productivity and decrease costs of operations, using a series of management practices developed first in Japan and then adapted to worldwide circumstances. Rapid technological progress has opened new business potentials and opportunities forcing companies to constantly introduce more and more advanced solutions to remain competitive. However, the potential of such technologies to support the implementation of Lean Manufacturing is not completely perceived yet. After defining Lean 4.0 which aims to provide useful insight for the integration of lean and Industry 4.0 in the manufacturing companies, AI is then briefly introduced, as a key enabler. This paper focuses on the integration and support of AI to Lean 4.0, mainly investigating the process and control issues through better use of data and knowledge.
HAL (Le Centre pour la Communication Scientifique Directe), Feb 13, 2023
HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific r... more HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
Springer eBooks, 1996
This paper presents a new method of texture characterization and defect detection in textures, ba... more This paper presents a new method of texture characterization and defect detection in textures, based on mathematical morphology transformations: structural opening and top-hat. The structural opening and its properties of invariance allow the extraction of primitive patterns from a texture, some kind of “textons” which entirely characterize it. As a result, a defect detection can be performed by a top-hat transformation for a similar textural set with defects.
ABSTRACT This paper presents a new method of texture characterization, recognition and defect det... more ABSTRACT This paper presents a new method of texture characterization, recognition and defect detection in textures, based on mathematical morphology transformations: structural opening and top-hat. The structural opening and its properties of invariance allow the extraction of primitive patterns from a texture, some kind of “textons” which entirely characterize it. As a result, defect detection can be performed by a top-hat transformation for a similar textural set with defects, and texture recognition can be led by a reference set
arXiv (Cornell University), Sep 30, 2022
his paper reviews the entire engineering process of trustworthy Machine Learning (ML) algorithms ... more his paper reviews the entire engineering process of trustworthy Machine Learning (ML) algorithms designed to equip critical systems with advanced analytics and decision functions. We start from the fundamental principles of ML and describe the core elements conditioning its trust, particularly through its design: namely domain specification, data engineering, design of the ML algorithms, their implementation, evaluation and deployment. The latter components are organized in an unique framework for the design of trusted ML systems.
HAL (Le Centre pour la Communication Scientifique Directe), Jul 2, 2018
Avec la transformation numérique du Groupe Thales, l'intelligence artificielle (IA) qu'elle soit ... more Avec la transformation numérique du Groupe Thales, l'intelligence artificielle (IA) qu'elle soit orientée par les modèles (IA symbolique) ou basée sur les données (IA connexioniste) devient un élément clé pour la maintenance prévisionnelle. L'objectif de cet article est de dresser un rapide panorama des différentes technologies d'IA mise en oeuvre dans Thales pour outiller les activités de maintenance prévisionnelle.
Uploads
Papers by Juliette MATTIOLI