Search results

Japan is almost 10,000 kilometers from Belgium, a country that fascinates, not least for its rich culture full of contrasts. Researchers at UNamur maintain close ties with several Japanese institutions, particularly in the fields of computer science, mathematics and video games. Let's take a look at some of these collaborations..

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

Can we still speak of democracy when algorithms influence our electoral choices or participate in the drafting of laws? This topic is explored by Aline Nardi, researcher at the Faculty of Law and member of the Namur Digital Institute (NADI).

The Namur Digital Institute (NaDI) is launching a series of original events: "Les Séances du Numérique". Films followed by debates with experts to understand digital challenges and stimulate collective thinking. A project spearheaded by Anthony Simonofski, Anne-Sophie Collard, Benoît Vanderose and Fanny Barnabé.

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

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

Trust of traditional political institutions and elected representatives, rise of authoritarian logics, definition of public services... Democracy today seems to be going through a turbulent zone. What role does the university play in this context? To shed light on this question, we interviewed four researchers from different disciplines: educationalist Sephora Boucenna, philosopher Louis Carré, political scientist Vincent Jacquet and legal scholar Aline Nardi. Their contrasting views sketch out the contours of an issue that is more topical than ever: thinking about and defending the link between university and democracy.

The IRDENa institute invites its members and guests to present the status of their research in teaching and education on topical issues throughout the year.These seminars are held once a month, usually during lunchtime. We look forward to seeing many of you at these events, which provide an opportunity for discussion, debate, and sharing in a friendly atmosphere.

On May 12, the Institute for Research in Didactics and Education (IRDENa) at the University of Namur is organizing a seminar dedicated to a topic at the heart of current concerns in initial and continuing teacher education: training through professional practice.In a context where expectations of the teaching profession and the realities on the ground are evolving rapidly, and where policymakers are entrusting the field with a significant portion of the training of future teachers through the requirement for extended teaching placements, practical experience plays a decisive role in the development of professional competencies, professional skills, and educational approaches. This essential practice also raises numerous questions:What challenges does training in (with and through) professional practice face today?What obstacles still hinder its implementation or quality?What challenges must training institutions, partner schools, and trainers address?What concrete benefits does professional immersion offer for both aspiring and experienced teachers?To shed light on these questions, the event will feature two speakers:Catherine Van Nieuwenhoven, professor and international expert on teacher education and work-study programs;Sephora Boucenna, a researcher at IRDENa, whose work focuses on professional development through practice and the analysis of field experiences.Their combined perspectives, blending scientific expertise, institutional analysis, and a nuanced understanding of the field, will fuel a collective discussion on the levers to strengthen and the avenues to explore in order to support an ambitious, coherent, and efficient practical training program in light of current policy challenges.The program will continue with a roundtable bringing together colleagues from various universities, who will compare their perspectives on the training of internship supervisors. Their discussion will focus in particular on training needs, support models, institutional challenges, and the conditions necessary to organize this training effectively.Finally, the day will give significant attention to real-world practice through testimonials from several student teachers, who have been invited to share their experiences, their mentoring practices, as well as the tensions that arise in their daily professional lives. These accounts will help ground the discussion in the concrete realities of schools and highlight the importance of the partnership between academic institutions and K-12 education.This study day is intended for researchers, trainers, teachers, as well as anyone involved in teacher education who wishes to contribute to a shared reflection on the future of the profession. Program 8:30–9:00 a.m. – Welcome9:00 AM to 10:00 AM – Catherine Van Niewenhoven (UCLouvain) - The Role of Fieldwork in Teacher Education: Support and Professional Development10:00 AM to 10:20 AM - Coffee break10:20 AM to 11:20 AM – Research findings (Call for papers)11:20 AM to 12:20 PM - Presentation of mentor programs12:30 PM to 1:30 PM – Lunch1:30 PM to 2:00 PM - Research Incubator2:00 PM to 3:00 PM - Sephora Boucenna (UNamur) - Training through and within professional practice: what are the specific features?3:00 PM to 4:00 PM - RoundtableStarting at 4:00 PM - Closing reception

Flamure Ibrahimi is a Ph.D. student in service and marketing management at the NaDI-CeRCLe Research Center at the University of Namur (Belgium) within the EMCP Faculty, under the supervision of Prof. Dr. Wafa Hammedi (University of Namur) and Prof. Dr. Linda Alkire (Texas State University). She has just been awarded the prestigious ServCollab Scholarship 2026, an international distinction that recognizes and supports doctoral students whose work falls within the field of Transformative Service Research (TSR)—doctoral projects with high impact on society and humanity.