Search results

Abstract Estrogens originating from human and animal excretion, as well as from anthropogenic sources such as cosmetics, plastics, pesticides, detergents, and pharmaceuticals, are among the most concerning endocrine-disrupting compounds in aquatic environments due to their potent estrogenic activity. While their effects on fish reproduction are well documented, their impact on development, particularly metamorphosis, remains poorly studied. This hormonal transition, mainly controlled by the thyroid axis, is essential for the shift from the larval to the juvenile stage in teleosts.The effects of two contraceptive estrogens on zebrafish (Danio rerio) metamorphosis were evaluated: 17α-ethinylestradiol (EE2), a synthetic reference estrogen, and estetrol (E4), a natural estrogen recently introduced in a new combined oral contraceptive formulation. Continuous exposure from fertilization to the end of metamorphosis allowed the assessment of morphological changes, disruptions of the thyroid axis, and modifications of additional molecular pathways potentially involved in metamorphic regulation.EE2 induced significant delays and disturbances in metamorphosis, affecting both internal and external morphological traits, confirming its role as an endocrine disruptor of concern. In contrast, E4 did not cause any detectable morphological alterations even at concentrations far exceeding those expected in the environment, indicating a limited ecotoxicological risk. Molecular analyses showed that EE2 strongly affected thyroid signaling and energy metabolism during metamorphosis, whereas E4 induced only minor transcriptional and proteomic changes.This study provides the first evidence that EE2 can disrupt zebrafish metamorphosis and highlights the importance of including this developmental stage in ecotoxicological assessments. The results also suggest a larger environmental safety margin for E4, although further research is needed to clarify the mechanisms linking estrogen exposure to metamorphic regulation.JuryProf. Frederik DE LAENDER (UNamur), PresidentProf. Patrick KESTEMONT (UNamur), SecretaryDr. Sébastien BAEKELANDT (UNamur)Dr. Valérie CORNET (UNamur)Prof. Jean-Baptiste FINI (Muséum National d'Histoire Naturelle, Paris)Dr. Marc MULLER (ULiège)Prof. Veerle DARRAS (KULeuven)

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

A flurry of awards for the NaDI-CeRCLe research center in recent weeks. The service management research of three young doctoral students from the EMCP Faculty has been recognized by their peers at leading international scientific events: Floriane Goosse, Victor Sluÿters and Florence Nizette. This summer, we invite you to discover their careers and their work.

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..