Seminars

by Dr Dennis Soemers, Department of Advanced Computing Sciences of Maastricht UniversityResearchers in artificial intelligence (AI) often seek to develop algorithms that exhibit high degrees of generality, as well as generalisation. "Generality" refers to the ability to apply the same technique to multiple different problems. "Generalisation" refers to the ability of a system to generalise effectively from observed situations to unseen ones. In this talk, I will argue that the research community currently focuses too much on techniques that are generally applicable and can generalise well in principle, but that delivering on these promises by truly scaling up to wide varieties of problems becomes infeasible in practice. Paradoxically, reducing the scope of problems that can be tackled by any single system in theory, may dramatically increase the ease with which it can be deployed to the remaining scope in practice. Pay attention : Sandwiches will be provided, please fill in this form before day D at 09:00 to reserve a sandw. Practical information : March 20th 2025;  1:00-2:00 pm;  Shannon room - Maxwell, a.105 - Place du Levant, 3

Information security and data protection are at the core of architects and developers during the design and development of applications with users (web or mobile applications, ATMs, embarked software, etc.). In order to prevent any unforeseen usage of these applications, professional hackers are involved to detect any breach that could be used for fraudulent usage of these applications.Based on multiple years of practical experience, the speaker will introduce and illustrate methodologies and tools that are commonly used by professional hackers.Seminar given as part of the LINFO2399 - Industrial Seminar in Computer Science

ICTEAM hosts a Francqui chair holder, Prof. Hannu Toivonen, from the University of Helsinki, Finland. Prof. Toivonen is a world-renowned expert in data mining and computational creativity. On this occasion, we are delighted to organizea half-day seminar on computational creativity. We will explore how AI can generate, enhance, or collaborate in creative processes, producing novel outputs in various domains such as music and games.Professor Siegfried Nijssen has the pleasure to welcome the following experts:• Hannu Toivonen, Professor of Computer Science at the Department of Computer Science of the University of Helsinki, Finland. He will talk about Creative Computers – An Oxymoron?• Peter Van Roy, Professor of Computing Science and Engineering, ICTEAM, EPL, UCLouvain. He will talk about  Music composition aided by symbolic AI• Geraint A. Wiggins, Professor of Computational Creativity at the VUB and at Queen Mary University of London, UK. He will talk about  Information Dynamics, Spectral Knowledge Representation, and the simulation of music cognition• Eric Piette,  Professor  of Computing Science and Engineering, ICTEAM, EPL, UCLouvain. He will talk about  Generating Games Via Evolution and Language ModelsThe presentations will be followed with a discussion and a drink.This  event is a great opportunity for PhD students, researchers, and faculty to explore cutting-edge AI applications in creativity and share insights.  

by Jing-Rebecca Li (INRIA-ENSTA)The complex transverse water proton magnetization subject to diffusion-encoding magnetic field gradient pulses in a heterogeneous medium such as brain tissue can be modeled by the Bloch-Torrey partial differential equation. The spatial integral of the solution of this equation in realistic geometry provides a gold-standard reference model for the diffusion MRI signal arising from different tissue micro-structures of interest. A closed form representation of this reference diffusion MRI signal called matrix formalism, which makes explicit the link between the Laplace eigenvalues and eigenfunctions of the biological cell and its diffusion MRI signal, was derived 20 years ago. In addition, once the Laplace eigendecomposition has been computed and saved, the diffusion MRI signal can be calculated for arbitrary diffusion-encoding sequences and b-values at negligible additional cost. Up to now, this representation, though mathematically elegant, has not been often used as a practical model of the diffusion MRI signal, due to the difficulties of calculating the Laplace eigendecomposition in complicated geometries. We present a simulation framework that we have implemented inside the MATLAB-based diffusion MRI simulator SpinDoctor that efficiently computes the matrix formalism representation for realistic neurons using the finite element method. We show that the matrix formalism representation requires a few hundred eigenmodes to match the reference signal computed by solving the Bloch-Torrey equation when the cell geometry originates from realistic neurons. As expected, the number of eigenmodes required to match the reference signal increases with smaller diffusion time and higher b-values. We also convert the eigenvalues to a length scale and illustrate the link between the length scale and the oscillation frequency of the eigenmode in the cell geometry.