Scientific Publications of the University of Toulouse II Le Mirail
Not a member yet
92205 research outputs found
Sort by
Autonomous Vehicles: safe and dangerous all at once: Understanding the social representations of Autonomous Vehicles for safer roads
No full textInternational audienc
La face cachée des techniques de procréation
No full textFiche technique Auteur(s) : Anne-Sophie Giraud Interview : Yulia Potié Producteur : Université Toulouse – Jean Jaurès / MIN / Mondes Sociaux Réalisateur(s) : Cédric Peyronnet Éditeur : Pôle Production – Le vidéographe / Université Toulouse – Jean Jaurès Date de réalisation : 08/01/2025 Durée du programme : 10 min 23 Langue : Français Conditions d’utilisation : Tous droits réservés à Université Toulouse – Jean Jaurès et aux auteurs.Devenues populaires et très médiatisées, les techniques procréatives comme la fécondation in vitro ou le diagnostic préimplantatoire sont souvent sur le devant de la scène. Solution miracle pour certains et problème éthique pour d’autres, ces méthodes font débat mais que sait-on réellement sur ces techniques et sur ce que vivent les personnes qui y ont recours ?Dans ce nouvel épisode, l’anthropologue Anne-Sophie Giraud nous emmène à la découverte des techniques procréatives et de la réalité parfois difficile pour les personnes qui suivent ces parcours.Auteur(s) : Anne-Sophie GiraudInterview : Yulia PotiéProducteur : Université Toulouse – Jean Jaurès / MIN / Mondes SociauxRéalisateur(s) : Cédric PeyronnetÉditeur : Pôle Production – Le vidéographe / Université Toulouse – Jean Jaurè
Descriptor System Modeling for Robustness in Control
No full textInternational audienceThe robust control framework considers systems regulated close to an equilibrium point or a trajectory. Under these assumptions the systems are linear but dependent of many intricate features such as uncertainties, neglected dynamics, time-dependent parameters, isolated non-linearities, time-delays... Robustness is achieved when some performance can be assessed even for the worst configuration of these features. Mathematically, it boils down to proving that some complicated non-linear function of many indeterminates is negative when these indeterminates are constrained in sets that are themselves described by non-linear equations. In this talk we show that for many cases the Descriptor System Modeling that combines Algebraic and Differential Equations is highly useful to simplify (linearize) the problems. Combined to the S-variable approach which cleverly exploits Finsler's lemma, it provides easy to code Semi-Definite Programming relaxations. These numerically tractable relaxations are conservative, yet Descriptor-based liftings of the model allows to build hierarchies of relaxations with decreasing conservatism. In the talk we will give an insight on these theoretical results illustrated on some simple examples
Alignment of Schema-Only and Instance-Only Data Sources Using Large Language Models
No full textInternational audienceData matching is a persistent challenge in heterogeneous data integration. Traditional data matching methods, relying on schemabased, instance-based, or hybrid approaches, often fall short when aligning disparate data from schema-only with instance-only sources. To address this problem of aligning disparate sources, we present an innovative data matching framework that enables the matching of one data source, where only schema-related information is available, with another data source, where only instance-related information is available. The strength of our framework lies in its ability to combine outputs from multiple Schema-Instances matchers, generating auxiliary information to enhance the alignment between disparate data structures. Our framework is validated using the Valentine Benchmark through extensive experiments. These findings underscore the potential of our approach to advance the integration of diverse data sources.</div
Convergence rates for polynomial optimization on set products
No full text22 pagesInternational audienceWe consider polynomial optimization problems on Cartesian products of basic compact semialgebraic sets. The solution of such problems can be approximated as closely as desired by hierarchies of semidefinite programming relaxations, based on classical sums of squares certificates due to Putinar and Schm\"udgen. When the feasible set is the bi-sphere, i.e., the Cartesian product of two unit spheres, we show that the hierarchies based on the Schm\"udgen-type certificates converge to the global minimum of the objective polynomial at a rate in , where is the relaxation order. Our proof is based on the polynomial kernel method. We extend this result to arbitrary sphere products and give a general recipe to obtain convergence rates for polynomial optimization over products of distinct sets. Eventually, we rely on our results for the bi-sphere to analyze the speed of convergence of a semidefinite programming hierarchy approximating the order quantum Wasserstein distance
Experimenting with a formative evaluation method in grassroots innovation
No full textContext: An emerging scientific consensus argues that addressing societal grand challenges (e.g., the sustainable development goals) implies the need for systems transformation (Von Schomberg 2013; Cagnin, Amanatidou, and Keenan 2012; Kuhlmann and Rip 2014). Such transformations involve new policy rationales and innovation approaches. Among these transformations is the growing consensus that there is a need to include civil society organisations in research and innovation processes. This is referred to as grassroots innovation (Smith and Seyfang 2013). Grassroots is an umbrella concept for a wide range of civil society organisations (CSO) such as cooperatives, voluntary associations, informal community groups, voluntary labour, and the social economy. CSOs tackle significant social issues and experiment “creative and purposeful changes in social practices” (Howaldt & Kaletka, 2023: xxv). In that matter, they are strategic actors of social innovation. Questioning the transformative potential of grassroots innovation is tantamount to: 1. considering that it can transform society for the common good (Joly 2020); 2. adopting a systemic and dynamic approach to innovation (Geels and Schot 2007), and 3. following a social innovation initiative that takes account of the changing context and envisages societal impacts as a shared future. Research question: To which extent can formative evaluation foster social learning and steer grassroots innovation towards expected transformations? We address this question within the context of a transformative social innovation project. We conceived and carried out this project (called EQUIPACT) to produce knowledge to enhance the transformative capacity of grassroots innovation. EQUIPACT is part of a process supported by a multi-stakeholder platform, Alliss. For the past ten years, this platform has been working to highlight the societal impact of grassroots innovation and to legitimize CSOs in knowledge co-production processes. EQUIPACT is supported by a hybrid consortium representing the diversity of organisations involved in grassroots innovations. It is based on three principles: 1) the presence of a variety of expertise increases the transformative capacity of grassroots innovation; 2) change occurs because the actors involved learn from each other; and 3) the system in which it takes place is constantly changing. These principles imply changes in the governance, organisation and implementation of innovation. They also require the development and use of new approaches for assessing the impact of innovation. We choose to use a formative evaluation approach (Molas-Gallart et al. 2021; Matt et al. 2023) which allows learning while directing innovation towards societal impact. In innovation studies, there has been a robust stream of research focused on understanding how multi-stakeholder groups learn from their experiences so to implement and scale-up system innovations. This stream of research has been referred to as social learning (Guijt, Woodhill, and others 2002). Social learning scholars mention the importance of reflexivity when learning contexts are characterised by diverse values, interests and knowledge. Directing innovation towards societal impacts requires a different understanding and means of impact assessment (Matt et al., 2023).Methods: Societal impacts depend on the occurrence of a series of systemic changes that will facilitate their realization (Torfing, Sørensen, and Røiseland 2019). This is why we believe that we need to support stakeholders in co-steering a grassroots innovation project using a formative evaluation method. We have previously experimented using the ASIRPA Real-time (rt) method in cities for co-creating solutions to carbon emissions with residents (Matt et al. 2023; van Dis et al. 2023), and more recently, with food system actors co-innovating around agroecological crop protection. ASIRPArt has two advantages: 1) it takes account of the high degree of uncertainty and complexity encountered by the developers of such projects; and 2) it helps them to develop a collective vision of the future and to learn about systemic changes along the way. The ability to steer innovation processes in the desired direction is fundamental to achieving societal grand challenges (Grillitsch et al. 2019). Briefly, ASIRPArt was implemented through workshops where the members of the EQUIPACT consortium drew an impact pathway (Matt et al, 2023). This pathway charts the path to be followed to reach the goals the members have collectively set. It was used as the core tool to orient participants and guide their anticipatory actions during the project and beyond. The iterative process was achieved through yearly revisions of the impact pathway.Results: According to our working hypothesis, the ASIRPArt approach can improve reflexivity and build directionality in innovation ecosystems (van Dis et al. 2023). Reflexive revisions of the impact pathway enables the participants to check their progress towards their goal. If there are deviations, they can adopt corrective actions to either get back on the original impact pathway or to establish a new pathway. Since the EQUIPACT project unfolded over 2 years, the pathway was revised twice, the second revision occurring weeks prior to the closing symposium. We will present the social learning that emerged through the workshop participation, the changes partners observed within their organisations and beyond, and the actions they anticipated to continue the initiative following the closings of the EQUIPACT project. We can already conclude that social learning does not only depend on the iterative use of the ASIRPArt method within a consortium; it also depends on the organisations’ capacity for reflexivity.Dans cette communication, nous présenterons les résultats de l’expérimentation d’une méthode d’évaluation formative dans le cadre d’un projet de recherche-action dont l’objectif est d’équiper les acteurs de la recherche participative pour plus d’impacts dans les transitions (EQUIPACT) . S’interroger sur le potentiel transformatif de la recherche participative (RP) revient à : 1. considérer que la RP est à la recherche citoyenne ce que la recherche partenariale est ) la recherche industrielle : une alliance avec la recherche académique pour l’intérêt général (Joly, 2020) ; 2. adopter une approche systémique et dynamique de l’innovation (Geels et Schot, 2020), 3. Suivre un cap qui prenne en compte le contexte changeant et envisage les impacts sociétaux comme un futur commun. Ces impacts sociétaux dépendent de la survenue d’une série de changements systémiques qui en faciliteront la réalisation (Torfing et al., 2019). C’est pourquoi nous postulons qu’il faut accompagner les acteurs dans le co-pilotage d’un projet de recherche-action à visée transformative avec une méthode d’évaluation formative. Nous avons expérimenté la méthode ASIRPA temps réel avec des villes dans des processus de co-création avec les habitants (Matt et al, 2023, van Diss et al., 2024), et avec des parties prenantes dans des systèmes agroécologiques (Matt et al, 2024). ASIRPA temps réel présente deux avantages. Elle tient compte du haut degré d’incertitude et de complexité que rencontrent les porteurs de tels projets. Elle les aide à porter une vision collective du futur et à apprendre le changement « chemin faisant ». Ce projet s’inscrit dans une dynamique portée par une plateforme multi-acteurs ; Alliss. Depuis une dizaine d’années, cette plateforme a œuvré pour mettre en lumière les impacts sociétaux de la RP, faire reconnaitre le rôle du tiers-secteur dans les processus de co-production de connaissances. EQUIPACT est porté par un consortium hybride et représentatif de la diversité des organisations impliquées dans les processus de RP. Il est coordonné par le LISIS et une association Le Dome, à Caen. EQUIPACT repose sur trois principes : la présence d’une variété d’expertises augmente la capacité transformative d’une recherche-action ; les changements surviennent parce que les acteurs impliqués apprennent les uns des autres, le système dans lequel il s’inscrit, se transforme constamment. Ces principes impliquent de changer la gouvernance, l’organisation, et la réalisation de la recherche. C'est cette démarche que nous présentons dans cet article
Les représentations savantes de l’architecture rurale par les enquêtes du M.N.A.T.P. de 1941 à 2001
No full textInternational audienc
Retour au lac d'Issarlès : un premier bilan d'étape
No full textInternational audienceLe lac d'Issarlès, joyau touristique du volcanisme se situe dans le très haut bassin de la Loire, dans l'ancienne province du Vivarais. C'est le lac d'origine volcanique le plus profond de Franc
JC3DHN 2024 - Actes du colloque
No full textInternational audienceCet ouvrage rassemble les actes du colloque du Consortium 3DHN, qui s’est tenu du 26 au 28 novembre 2024 à Nancy. Le Consortium-HN-3DHN est un réseau national, labelisé par l’IR * Huma-Num, qui fédère des équipes de recherche ainsi que des institutions patrimoniales, œuvrant pour le développement, la structuration et la diffusion des bonnes pratiques liées à la production, la gestion, la conservation et la valorisation des données 3D dans le domaine des sciences humaines et sociales.À travers ses actions collectives, le consortium-HN-3DHN poursuit plusieurs objectifs majeurs : accompagner les chercheurs et les acteurs du patrimoine dans l’usage raisonné et critique des technologies 3D. favoriser l’interopérabilité des données 3D, promouvoir les principes FAIR et encourager l’adoption de standards ouverts.Le colloque de Nancy a constitué un moment privilégié d’échange et de réflexion sur ces thématiques et a permis de confronter les expériences, discuter des méthodologies, et d’explorer les perspectives offertes par les approches 3D au sein des Humanités Numériques.L’ensemble des interventions du colloque est consultable sur la chaîne du consortium 3DHN, hébergée sur la plateforme audiovisuelle de l’enseignement supérieur et de la recherche Canal-U (https://www.canal-u.tv/chaines/c3dhn
