Transfert de connaissances

Résumé

vignette|Logo illustratif de The Noun Project. Le transfert de connaissances ou compétences, dans les domaines du développement et de l’apprentissage de l'organisation, est le problème pratique de la transmission de données d’une partie de l’organisation à une autre (ou aux autres) partie(s). Le transfert de connaissances ne recouvre qu'une partie de la problématique du transfert de compétences pour les structures. Les entreprises ou organisations voulant sécuriser leurs compétences stratégiques rencontrent certaines difficultés à préserver et transmettre les savoir-faire rares, peu ou pas formalisés, tacites et uniquement basés sur l'expérience du salarié. S'il est vrai que le transfert de connaissances peut faire l'objet d'un enregistrement et d'une sauvegarde de l'ensemble des supports consignés des pratiques métier d'une organisation (consigne, mode opératoire, référentiel, film, vidéo, etc.), pour autant cela ne suffit pas à préserver les compétences clés de l'entreprise et encor

Source officielle

https://fr.wikipedia.org/wiki/Transfert_de_connaissances

À propos de ce résultat

Cette page est générée automatiquement et peut contenir des informations qui ne sont pas correctes, complètes, à jour ou pertinentes par rapport à votre recherche. Il en va de même pour toutes les autres pages de ce site. Veillez à vérifier les informations auprès des sources officielles de l'EPFL.

Publications associées (36)

Publications associées

Chargement

Personnes associées (2)

Personnes associées

Chargement

Unités associées (2)

Unités associées

Chargement

Concepts associés (9)

Concepts associés

Chargement

Cours associés (5)

Cours associés

Chargement

Séances de cours associées (2)

Séances de cours associées

Chargement

Economic growth is increasingly explained and driven by knowledge acquisition and firmsâ capabilities to adapt to new evolving situations to fit economic, societal and environmental frames. Strategic behavior of private companies in differentiating and promoting their products and services is crucial for their competitiveness. In parallel, public research performed at different territorial levels and by a high diversity of actors is a strategic source of key assets adding value to the firms. External actors operate in a non-linear structure allowing frequent feedback and interactive loops in order to optimise the replies to substantial problems and constantly valorise their resources. These features are explored in the thesis through different projects. This dissertation is motivated by international negotiations in the frame of the GATT agreements and bilaterally between the European Union and Switzerland that are going towards the breakdown of borders, threatening the competitiveness of the Swiss agriculture. Negative externalities will affect small firms composing this agricultural network. Thus, the dissertation is focusing on innovation as a strategic asset to maintain the competitiveness and differentiate the domestic agricultural products. The first research project concentrates on interactions occurring in the Swiss apricot network. We investigate the types of interactions producing innovations that come from different sources that are internal or external to the agricultural production supply chain. We use network theory to model the structure of the domain and understand what is inherent to innovation production thanks to collaborations through an empirical work. In the second research project, we examine whether there is a connection between the issues faced by small companies producing food and the innovative solutions and knowledge generated by research institutions. The importance of solutions to daily activities conducted by companies depends on the nature of the sector. Thus, research projects providing outputs like publications might not specifically target these issues. Moreover, access, awareness and affordability of innovations can be hindering factors for small firms composing a low-technology intensive sector. Gaps of knowledge transfer, innovation and implementation are identified in this setting. In the last research project, we address market failures hampering economic and strategic management decisions encountered in the agricultural sector. We empirically construct two case studies highlighting the necessity of intellectual appropriation mechanisms. We explore the use of patents and regional trademark to add value to the emblematic product of a small region. Organisational and marketing innovations have to be supported and diffused by key stakeholders, aiming a good communication based on criteria that take into account diverse interests of those stakeholders; consumers have to be involved in the innovation process. Furthermore, knowledge transfer, innovation generation, diffusion and implementation rely upon the will of stakeholders to take part in an efficient process of interactions and communication in order to produce and perform useful products and services. Practical implications are levered in the projects presented. Finally, digital technologies are efficient for information exchange and can decrease time and transaction costs. These technologies are explored as part of the solution to small firmsâ

EPFL2019

Chargement

In this dissertation, we propose gradient-based methods for characterizing model behaviour for the purposes of knowledge transfer and post-hoc model interpretation. Broadly, gradients capture the variation of some output feature of the model upon unit variation of an input feature, and thus encodes the local model behaviour while being agnostic to the underlying model architectural choices.Our first contribution is to propose a sample-efficient method to mimic the behaviour of a pre-trained teacher model with an untrained student model using gradient information. We interpret our approach as an efficient alternative to data augmentation used with canonical knowledge transfer approaches, where noise is added to the inputs. We apply this to distillation and a transfer learning task, where we show improved performance for small datasets.Our second contribution is to propose a novel saliency method to visualize the input features that are most relevant for predictions made by a given model. We first propose the full-gradient representation, which satisfies a property called completeness which provably cannot be satisfied by gradient-based saliency methods. Based on this, we propose an approximate saliency map representation called FullGrad which naturally captures the information within a model across feature hierarchies. Our experimental results show that FullGrad captures model behaviour better than other saliency methods.Our final contribution is to take a step back and ask why input-gradients are informative for standard neural network models in the first place, especially when their structure may as well be arbitrary. Our analysis here reveals that for a subset of gradient-based saliency maps, the map relies not on the underlying discriminative model p(y | x) but on a hidden density model p(x | y) implicit within softmax-based disciminative models. Thus we find that the reason input-gradients are informative is due to the alignment of the implicit density model with that of the ground truth density, which we verify experimentally.

EPFL2021

Chargement

Are the Innovation Needs of Low-Technological Small and Medium-Sized Enterprises in Line with Knowledge Production by Research Institutions?

Camille Marguerite Aouinait

Small agricultural firms have to cope with increasing economic, legal and environmental difficulties. The purpose of the paper is to identify whether there are available research outputs to these challenges, focusing on the Swiss traditional fruit sector. Identification of firms' needs using face-to-face interviews and a focus group allowed us to prioritize innovations. Knowledge and innovation transfer gaps were identified thanks to bibliometric analysis, providing hints for research program orientations. The main sectoral needs were related to varieties like increasing pest and disease resistance, weather sensitivity, and ripening behavior. Knowledge produced by research institutions was partly related to the firms' needs. The study demonstrated a need for better coordination between agricultural stakeholders and researchers. Knowledge must be transferred and shared in the network with greater efficiency by extending interactions and involving stakeholders earlier in the innovation process. However, the search for scientific outputs should include more specificities of the local environment.

DE BOECK UNIV2019

Chargement

EPFL2021

EPFL2019