Search results

A lifelong video game enthusiast, Fanny Barnabé, a researcher at the CRIDS research center (Namur Digital Institute) and lecturer at the University of Namur, explores behind the scenes of a major cultural phenomenon: video game streaming on Twitch. Between humor, irony and toxic discourse, she deciphers the issues at stake in a digital space in the throes of change.

High-Sensitivity Birefringence Mapping Using Near-Circularly Polarized Light I will describe several techniques for mapping a two-dimensional birefringence distribution, which can be classified according to the optical schemes and principles of work:Illumination geometry (transmitted light/reflected light)Image acquisition (sequential acquisition/simultaneous acquisition)Polarization control (electrically controlled variable retardance/mechanical rotation).This classification facilitates a comparative analysis of the capabilities and limitations in these methods for birefringence characterization. Polychromatic polarizing microscopy (PPM) provides unique capabilities to alternative methods. It leverages vector interference to generate vivid, full-spectrum colors at extremely low retardances, down to < 10 nm. PPM is a significant departure from conventional polarizing microscopes that rely on Newton interference, which requires retardances above 400 nm for color formation. Furthermore, PPM's color output directly reflects the orientation of the birefringent material, a feature absent in conventional microscopy where color is solely determined by retardance.Joint seminar of ILEE & NISM!The seminar is open to external people too, no need to register. More info

Abstract Deep learning has become an extremely important technology in numerous domains such as computer vision, natural language processing, and autonomous systems. As neural networks grow in size and complexity to meet the demands of these applications, the cost of designing and training efficient models continues to rise in computation and energy consumption. Neural Architecture Search (NAS) has emerged as a promising solution to automate the design of performant neural networks. However, conventional NAS methods often require evaluating thousands of architectures, making them extremely resource-intensive and environmentally costly.This thesis introduces a novel, energy-aware NAS pipeline that operates at the intersection of Software Engineering and Machine Learning. We present CNNGen, a domain-specific generator for convolutional architectures, combined with performance and energy predictors to drastically reduce the number of architectures that need full training. These predictors are integrated into a multi-objective genetic algorithm (NSGA-II), enabling an efficient search for architectures that balance accuracy and energy consumption.Our approach explores a variety of prediction strategies, including sequence-based models, image-based representations, and deep metric learning, to estimate model quality from partial or symbolic representations. We validate our framework across three benchmark datasets, CIFAR-10, CIFAR-100, and Fashion-MNIST, demonstrating that it can produce results comparable to state-of-the-art architectures with significantly lower computational cost. By reducing the environmental footprint of NAS while maintaining high performance, this work contributes to the growing field of Green AI and highlights the value of predictive modelling in scalable and sustainable deep learning workflows. Jury Prof. Wim Vanhoof - University of Namur, BelgiumProf. Gilles Perrouin - University of Namur, BelgiumProf. Benoit Frénay - University of Namur, BelgiumProf. Pierre-Yves Schobbens - University of Namur, BelgiumProf. Clément Quinton - University of Lille, FranceProf. Paul Temple- University of Rennes, FranceProf. Schin'ichi Satoh - National Institute of Informatics, Japan

Deadlines Opening of abstract submissions and registrations: September 15, 2025Deadline to submit indicative title and summary: November 30, 2025Deadline for final abstract submissions: May 1, 2026Early bird registration deadline: March 1, 2026 More information on the website

Abstract Deep learning has become an extremely important technology in numerous domains such as computer vision, natural language processing, and autonomous systems. As neural networks grow in size and complexity to meet the demands of these applications, the cost of designing and training efficient models continues to rise in computation and energy consumption. Neural Architecture Search (NAS) has emerged as a promising solution to automate the design of performant neural networks. However, conventional NAS methods often require evaluating thousands of architectures, making them extremely resource-intensive and environmentally costly.This thesis introduces a novel, energy-aware NAS pipeline that operates at the intersection of Software Engineering and Machine Learning. We present CNNGen, a domain-specific generator for convolutional architectures, combined with performance and energy predictors to drastically reduce the number of architectures that need full training. These predictors are integrated into a multi-objective genetic algorithm (NSGA-II), enabling an efficient search for architectures that balance accuracy and energy consumption.Our approach explores a variety of prediction strategies, including sequence-based models, image-based representations, and deep metric learning, to estimate model quality from partial or symbolic representations. We validate our framework across three benchmark datasets, CIFAR-10, CIFAR-100, and Fashion-MNIST, demonstrating that it can produce results comparable to state-of-the-art architectures with significantly lower computational cost. By reducing the environmental footprint of NAS while maintaining high performance, this work contributes to the growing field of Green AI and highlights the value of predictive modelling in scalable and sustainable deep learning workflows. Jury Prof. Wim Vanhoof - University of Namur, BelgiumProf. Gilles Perrouin - University of Namur, BelgiumProf. Benoit Frénay - University of Namur, BelgiumProf. Pierre-Yves Schobbens - University of Namur, BelgiumProf. Clément Quinton - University of Lille, FranceProf. Paul Temple- University of Rennes, FranceProf. Schin'ichi Satoh - National Institute of Informatics, Japan

On December 11, 2024, the University of Reims-Champagne-Ardenne hosted the launch event for the ORION project, in which the University of Namur is a partner. This project, financed for 4 years by ERDF and INTERREG funds, aims to improve water management in the Val de Meuse while preserving the ecosystems of the Val de Meuse, a river running through France and Belgium.

Max Collinet, professor of geology at the Faculty of Science and researcher at the Institute of Life, Earth and Environment (ILEE), has just been awarded equipment funding (EQP) from the F.R.S - FNRS following calls whose results were published in December 2024.