This document presents the main research problems addressed during my PhD studies. All these researches are led inside the two teams DBWeb in Télécom ParisTech and LInK (Learning and Integration of Knowledge) in AgroParisTech, both located in Paris, and supervised by Pr. Jean-Louis Dessalles and Pr. Antoine Cornuéjols. My researches focus on learning theory both in the perspective of symbolic m...

The goal of knowledge transfer is to take advantage of previous training experience to solve related but new tasks. This paper tackles the issue of transfer of knowledge between radial basis function neural networks. We present some preliminary work illustrating how a neural network trained on one task (the source) can be used to assist in the synthesis of a new but similar task (the target).

Cross-media retrieval is a research hotspot in multimedia area, which aims to perform retrieval across different media types such as image and text. The performance of existing methods usually relies on labeled data for model training. However, cross-media data is very labor consuming to collect and label, so how to transfer valuable knowledge in existing data to new data is a key problem towar...

Knowledge Transfer (KT) is a major part of technical aspect in space science. By involving the research council funded Knowledge Transfer Account (KTA) programme and the university industrial and commercial partners, these strengthen the future research collaboration amongst the university, industrial, commercial, scientific and engineering communities. Since the space science research and deve...

Knowledge and management of such knowledge have been studied for some time now as it became clear that a company with an unstructured approach to corporate knowledge management was incapable of competing in this new environment. As such, management of intangible assets, such as knowledge, came to be perceived as an important tool for competition. Based on that, this article aims to identify the...

We propose a novel learning framework for object categorization with interactive semantic feedback. In this framework, a discriminative categorization model improves through humanguided iterative semantic feedbacks. Specifically, the model identifies the most helpful relational semantic queries to discriminatively refine the model. The user feedback on whether the pattern is valid or not is inc...

This paper reports on a specific food and agribusiness industry project, employing new technological capabilities to better transfer expert knowledge. Knowledge transfer and technical support are key components of this project. VisIT, which stands for Visualization of Information Tool, is a powerful organizational tool, which provides an Internet-based graphical learning environment. As a means...

Classical distillation methods transfer representations from a “teacher” neural network to a “student” network by matching their output activations. Recent methods also match the Jacobians, or the gradient of output activations with the input. However, this involves making some ad hoc decisions, in particular, the choice of the loss function. In this paper, we first establish an equivalence bet...

Reusable experience packages aim to promote and diffuse knowledge rapidly and economically. The distribution can take place within the organization that produced the experience package or it may allow exporting knowledge from the organization producing the package to other ones [1, 2, 3]. This work analyses an experience related to knowledge transfer from a University research lab (Software Eng...