DISCO-Func: Discrete Optimization in Neural Network Function Classes through Submodular Analysis KU Leuven
Driven by advances in deep learning, powerful automated algorithms enable us to hold conversations with virtual assistants running on our mobile phone, cars are autonomously driving on streets in mixed traffic, increasingly sophisticated medical diagnostic software is assisting doctors in making life-saving decisions, and social networks and media libraries are automatically indexed and personalized to individual users. At the core of these ...
Secure and High-Performance Inter-Container Communication and Orchestration of Virtual Network Functions KU Leuven
Many industries and companies are migrating to a virtualized and orchestrated environment, allowing the deployment of Virtual Network Functions (VNFs) on cloud infrastructure. As a result, there is increasing popularity of the container orchestration frameworks for automating, scaling and managing high-availability applications towards the edge. With this trend, the security of these frameworks has become increasingly important, yet also ...
Generation of a human interactome network using MAPPIT Ghent University
Protein-protein interactions are crucial for virtually all biological processes. Detailed interaction networks can provide novel insights in the normal and pathological functions of a given protein. Analysis of the human interactome is done using the yeast two-hybrid technique, whereby the accuracy is sometimes questioned. We will contribute to these large-scales programs by asystematic analysis of observed interactions using MAPPIT.
Functionalization of Graphene with Physisorbed Self-assembled Networks for Electronic Applications KU Leuven
Graphene is a promising material for future electronic applications. However, a prerequisite for the use of graphene in several electronic applications is the controlled and reproducible modification of its electronic and chemical properties. Among the different strategies for surface functionalization, physisorption of molecular adsorbates has as significant advance that it does not result in degradation of the charge carrier mobility. ...
Over the last decades, the Internet has evolved from a research network connecting a couple of hundred nodes to a system connecting billions of users around the globe. This astonishing evolution has been driven by a continuous innovation, both in terms of technology, service offering and user requirements. While in the early days the Internet was mainly used for static services such as e-mail and web browsing, a wide variety of rich services ...
Algorithms for Efficient Management of Large-Scale Virtualized Networks by means of Advanced Learning and Reasoning Techniques Ghent University
The promising concept of geometric routing in large scale telecommunication networks will be extended to accommodate simultaneous embedding construction and dynamic assignment of virtual network functions. This requires extensive research into novel methods that tightly integrate learning and reasoning, using current and historical network and traffic information and capturing network dynamics via temporal dependencies.
Robust and energy-efficient virtual sensor networks. University of Antwerp
Intelligent Dense And Longe range IoT networks (IDEAL-IoT). University of Antwerp
We plan to diversify instruction set architectures by means of physically unclonable functions. Through compiler technology and virtual machines to be developed in this project, we will develop support for generating and running software on the diversified architectures. Each processor and each application for each processor then become unique, because of which they become protected against reverse-engineering and tampering.