Informatics at the service of collective well-being and personal development.

The Faculty of Informatics has a vision of a world in which the intensive and growing use of informatics is oriented towards collective well-being and personal development. Its mission is to contribute to this orientation through education, scientific research and service to society.

The studies

When you choose to study computer science, you're guaranteed immediate access to an exciting, multi-faceted career. Through a comprehensive range of courses - bachelor's, master's, specialization master's and doctorate - the Faculty of Computer Science offers you quality training based on scientific rigor and critical thinking, while emphasizing interdisciplinarity and societal responsibility.

Informatique études

Research

The mission of the Faculty of Computer Science is to ensure cutting-edge scientific research, open to the creation and integration of knowledge that feeds teaching, technological innovation and societal debate.

Faculté d'informatique robots

Service to society

The Faculty of Computer Science contributes to the development of our social, economic, technological and political environment by disseminating knowledge and providing advice at regional, national and international levels. Through the synergy between researchers and industry, and by making knowledge and know-how available, it participates in numerous missions of service to society.

Faculté d'informatique service

International

The Faculty of Computer Science's international outlook is fundamental to the development of its teaching and research activities.

Etudiants de l'UNamur

Organization

The Faculty of Computer Science has organized itself to manage its missions in the best possible way. It has around a hundred members at the service of teaching, research and service to society.

A word from the Dean

More portraits
IT is a key to understanding, progress and responsibility in the face of the challenges and issues facing our society.
Anthony Cleve
Dean of the Faculty of Computer Science
See content

Animation

A series of events and players revolve around the Faculty of Computer Science.

Animation étudiante fac info

Spotlight

News

30 Years of Part-Time Education: When Adults Become Students, Too

Law
Economy
Management
Marketing
Computer science (IT)
Students

For more than 30 years, UNamur has offered flexible-schedule programs, allowing people already in the workforce to enroll in bachelor’s, master’s, and specialized master’s degree programs. These programs are based on a tailored teaching approach centered on the students’ experiences and real-world situations. 

Etudiante en reprise d'études travaillant sur son ordinateur portable

For as long as he can remember, Théophile Renier, 31, has always wanted to be a lawyer. It’s a calling that this freelance music industry professional—who is also a marine technician—set aside for a time... before returning to it a few years ago by enrolling in the School of Law. Since 2016, UNamur has indeed offered a bachelor’s and master’s program in law with a flexible schedule.  “What I find exceptional about this program is that the professors actively support students in their success,” says Théophile. “Everything is set up so that people who can’t be there in person can still access the courses remotely. ” All classes in the flexible-schedule law program are recorded and posted online to facilitate remote learning. Students are also notified of assignment and exam deadlines well in advance, to accommodate their professional and family commitments. 

As Mathieu Rolain, the program’s academic coordinator, points out, students in the flexible-schedule law program stand out for their exceptional determination and discipline: “superheroes,” he says. So, despite working three days a week at the Port of Zeebrugge and running his own music business—“I work about 50 hours a week”—Théophile manages to juggle it all. “For me, it’s actually very complementary,” says this passionate student. “People often wonder if it’s harder to study when you already have professional experience, but for me, studying is a pleasure! Nowadays, I focus more on what’s essential.” After completing his first two years of his bachelor’s degree, the student is now starting his third year; he then plans to pursue a master’s degree before joining the bar. “For me, it’s a childhood dream coming true,” he says. The future lawyer has just one piece of advice for those who are still hesitating: 

Many people wonder if it’s possible, or if they’re too old: in my opinion, you should put your reservations aside! This program is truly designed for people who are currently working: it has everything you need to succeed. 

Théophile Renier Law student

Soft Skills and Skills Assessment

For the past 30 years, the EMCP Faculty (Economics, Management, Communication, Sciences Po) has also offered flexible-schedule programs. It was in this context that the FSE+ track “Assessment, Strengthening, and Enhancement of Skills,” led by Jean-Yves Gnabo, was developed. Thanks to external support from the European Social Fund Plus, this educational program now provides better support for adults returning to school, not only at EMCP but also, eventually, in other faculties at UNamur. The project’s uniqueness lies in its integration of two complementary dimensions. On the one hand, an academic framework centered on an innovative thesis format, designed to prevent students from dropping out near the end of the program due to “writer’s block.” On the other hand, there is support for transferable skills: communication, organization, leadership, teamwork, time management, stress management, and more.

“Historically, we’ve focused on subject-specific skills, while soft skills were assumed to be acquired,” explains Jean-Yves Gnabo. “But succeeding in a part-time program also depends on ‘soft’ skills.” Furthermore, there was another blind spot in our approach: the fact that students were not always aware of their own skills. ” The FSE+ program therefore allows students, with the help of a coach, to conduct a skills assessment upon entry and upon completion, thereby measuring their progress. “This is not only essential for self-confidence but also for learning how to leverage newly acquired skills in the professional world,” explains the economics professor. For the past two years, the EMCP Faculty has also offered a specialized master’s program with a flexible schedule focused on management and sustainable economics. Thirty-five students were already enrolled in the first year. “Once the master’s program is established, the idea is to draw on a board of external stakeholders facing transition challenges, as well as to build a community among students, alumni, and faculty,” concludes Jean-Yves Gnabo. 

IT Work Shifts: Training That's More Necessary Than Ever

For the past 25 years, it has also been possible to earn a master’s degree in computer science at UNamur through a part-time program. While the sector faces many questions related to the widespread adoption of AI, Vincent Englebert, a professor of computer science, has no doubt that this program will continue to attract many working professionals. “We’re obviously going through a turbulent phase, but one thing is certain: while AI boosts the number of lines of code we can write per day, it also introduces a fair amount of errors into those lines, which creates serious maintenance issues... Companies are therefore beginning to realize the ‘technical debt’ generated by AI: if we ignore an engine problem just because a car is still running, we run the risk of the vehicle exploding one day on the highway... ” For Vincent Englebert, training new professionals in the challenges of generative AI is therefore a necessity. “I’d even say that AI is a great opportunity to pursue a degree in computer science today! Thirty years ago, computer scientists were asked to develop very basic programs. Today, they’re asked to develop much more complex systems, and AI—when used properly—frees them from highly technical tasks in this context. This allows them to focus on what really matters: what humans want to achieve with AI to make it more ethical, fairer, more useful, and safer.” 

Master 60 en sciences économiques à horaire décalé

A teaching approach tailored to students' needs

At UNamur, adult education plays a key role. Some of these working adults return to university to advance their careers, while others do so because they are in the process of changing careers. They all arrive with their own experiences, expectations, and dreams, but also with family and professional responsibilities, time constraints, and doubts… That is why, through its various flexible-schedule programs (evening and Saturday classes), UNamur has developed a specific pedagogical approach to best support these adults in their new academic and professional ambitions: realistic time management, an emphasis on collaborative work, close interaction with faculty, and a co-constructive approach to learning...

This article is taken from the "Guest" section of Omalius magazine, Issue #42 (September 2026).

Omalius-Septembre-2026

A study by VUB and UNamur shows that AI agents develop their own “artificial natural language” to communicate about complex data

Artificial Intelligences (AI)

AI systems spontaneously develop their own language, tailored to their sensors, rather than copying human language.

IA

Researchers from the VUB’s Artificial Intelligence Lab and the University of Namur (UNamur) demonstrate, in a new large-scale study, how autonomous AI systems can spontaneously develop, without human intervention, a functional and shared language. Unlike large language models such as ChatGPT, which learn existing human language by heart, these AI agents construct their own communication system, directly rooted in their specific sensors and environment. This discovery represents a major breakthrough for communication between autonomous systems, such as self-driving cars, robots and industrial sensors.

A layer of abstraction for non-human senses

Current AI language models focus on understanding and generating human languages such as Dutch or English. These human languages are historically rooted in our own sensory perceptions. When physical or virtual AI agents are equipped with sensors for which humans have no direct sensory equivalent – such as lidar, sonar, infrared or chemical analysis – existing human language proves insufficient.

Jérôme Botoko Ekila, a PhD student at the VUB’s AI Lab, explains: ‘Instead of forcing AI systems to communicate in a human language that does not correspond to their sensors, we allow them to develop their own “artificial natural language”. They construct effective concepts that correspond exactly to what they perceive via their sensors, even if the data is incomprehensible to humans.’

As, in practice, the sensors on different machines or vehicles can never be calibrated identically, the researchers believe that this language offers the ideal solution.

Paul Van Eecke, professor of artificial intelligence at the VUB: “In industrial or physical applications, it is simply unrealistic to expect all the sensors on different robots or vehicles to be calibrated in exactly the same way. Humans solve this problem through language: if I say I am standing near the ‘red car’, we understand each other without our eyes registering exactly the same colour values. Language provides a layer of abstraction above the sensor readings, and this is precisely what we are now allowing these AI agents to construct autonomously.”

Self-organisation and evolution

Instead of pre-programmed words or fixed codes, the researchers simply provided the agents with the basic mechanisms underpinning the evolution of human language: the local invention of words, the adaptation of conceptual prototypes, and the incorporation of feedback following successful or failed communication.

Image
Katrien Beuls

What’s fascinating is that there’s no central computer imposing the rules. Each time, we let two agents communicate with one another at a local level. It is solely through these individual interactions and updates that a global convention emerges spontaneously across the entire population. This principle of self-organisation allows the language to optimise itself automatically: the agents create exactly the number of categories needed to carry out their tasks efficiently, without superfluous details.

Katrien Beuls Professor in the School of Computer Science and a member of the NaDI Institute at UNamur

Tested on 37 datasets

The universal applicability of the method has been demonstrated across 37 varied and complex datasets. The agents successfully developed their own vocabulary to communicate, notably regarding physico-chemical analyses of wine, NASA data on exoplanets, visual datasets, and images of birds or faces. In all these situations, the agents successfully synthesised the data into new, shared concepts.

Transparent and adaptive

Transparency is a notable advantage of the systems developed. Unlike many ‘black-box’ AI models, researchers can precisely determine the meaning of a specifically developed word, as the prototype can be directly linked to the values of the underlying sensors.

Furthermore, these evolving linguistic systems are extremely robust. When the environment changes, when sensors produce noise, or when individual agents have a different structure, the language automatically adapts to the new circumstances through continuous interaction.

Reference

Botoko Ekila, J., Verheyen, L., Nevens, J., Beuls, K., & Van Eecke, P. (2026). Emergent Communication in Continuous Worlds: Self-Organization of Conceptually Grounded Vocabularies at Scale. Transactions of the Association for Computational Linguistics, vol. 14, pp. 2234–2257. DOI: 10.1162/TACL.a.799

Funding and collaboration

This research was carried out as part of a collaboration between the Artificial Intelligence Laboratory at the Vrije Universiteit Brussel (VUB) and the Faculty of Computer Science at the University of Namur (UNamur), and was funded by the Research Foundation – Flanders (FWO) and the Belgian National Research Fund (FNRS) as part of the WEAVE project.

FNRS-FWO

Research from UNamur Makes Artificial Intelligence Tools More Reliable

Artificial Intelligences (AI)
Computer science (IT)
Digital transition

Researchers at UNamur have developed a new method that has uncovered 32 previously undetected bugs in some of the most widely used software libraries for training artificial intelligence models. Originating from a master’s thesis, this research has already led to fixes in several software programs used by millions of developers around the world. 

Une recherche de l'UNamur rend les outils de l'intelligence artificielle plus fiables

Before artificial intelligence can recognize an image, understand a text, or answer a question, it must first be trained. To do this, developers and data scientists use specialized software libraries—true “toolboxes” that enable the construction of these AI models.

A research team comprising the University of Namur, the University of Passau (Germany), and TU Delft (Netherlands) has been studying this software. Their goal: to detect bugs that are particularly difficult to spot and that could cause a program to crash suddenly.

A New Method for Detecting Hidden Bugs

To verify that these libraries are functioning properly, researchers use software capable of automatically generating thousands of tests. The purpose of these tests is to push the programs to their limits in order to identify any potential errors. 

Until now, however, these tools had a major limitation: whenever a test caused a software library to malfunction, the software responsible for running the tests would also stop working. This made it impossible to pinpoint the exact source of the problem or report it to the developers. 

Researchers at UNamur have therefore devised a new method. Now, each test is run separately. If one of them causes a crash, it is isolated: the testing software continues to run, logs the error, and carries on with its analysis. This makes it possible to automatically detect bugs that previously went unnoticed. 

32 bugs discovered and already fixed

Using this new method, researchers analyzed 1,648 modules from the leading machine learning libraries and uncovered 32 previously unknown bugs. Several of these affected software that has become indispensable for training artificial intelligence models.

These issues were reported to the development teams. Several have already been acknowledged and fixed. “When a bug is fixed at this level, potentially thousands, or even millions, of users benefit,” notes Xavier Devroey, a professor in the Faculty of Computer Science at UNamur and a member of the PReCISE research center at the Namur Digital Institute (NaDI). Beyond these fixes, this research helps make the software on which so many artificial intelligence applications rely today more reliable.

Research that grew out of a thesis

This breakthrough originated from Lucas Berg’s master’s thesis, completed at UNamur under the supervision of Professor Xavier Devroey. As part of his research internship, Lucas Berg then joined Professor Annibale Panichella’s team at TU Delft to work on Pynguin, the open-source automated test generation tool developed at the University of Passau by Professor Gordon Fraser’s team, mentioned earlier. 

Initially, his goal was to improve this tool. By preventing crashes while discovering bugs, this first contribution subsequently caught the attention of researchers at the University of Passau, who were facing the same problem, leading to an international collaboration that would result in a joint scientific publication. 

Lucas Berg is currently pursuing a Ph.D. at the Faculty of Computer Science at UNamur under the supervision of Professors Wim Vanhoof and Xavier Devroey. The results of this research were presented at the IEEE International Conference on Software Testing, Verification, and Validation (ICST 2026), one of the most prestigious international conferences in the field of software testing.  

This research also received support from the Wallonia Public Service (SPW Research) and Wallonia-Brussels International. This recognition underscores the quality of this work and its significance to the international scientific community. 

The Namur Digital Institute (NaDI)

At NaDI, researchers develop innovative solutions to the new societal challenges posed by the digital revolution (eGov, eHealth, eServices, big data, etc.). Drawing from a variety of disciplines, the researchers combine their expertise in computer science, technology, ethics, law, management, and sociology. Comprising six research centers, the Namur Digital Institute offers unique multidisciplinary expertise in all areas of computer science, its applications, and its social impact.  

AI and Quality Control: When Humans Stay in Control

Artificial Intelligences (AI)
Computer science (IT)
Medicine

How can anomaly detection on a production line be improved? It was in response to a very concrete problem encountered at the company where he works that Arnaud Bougaham, with the support of his employer, decided to devote his thesis to this topic at the UNamur School of Computer Science. The goal? To develop an artificial intelligence model that assists operators in detecting anomalies on an industrial production line by reducing false alarms and making the diagnosis easier to understand. This approach also holds promise for deployment in the medical field. 

IA

Missing components, fragile solder joints, reversed polarity, scratches, or foreign objects such as a screw that falls onto the product: on a production line for automotive electronic components, such as the one at AISIN Europe, defects—even if they are rare—must be detected immediately and accurately. “My goal was to teach a computer to distinguish between what is normal and what is abnormal in images,” explains Arnaud Bougaham, a data scientist at AISIN Europe and an associate researcher at the Faculty of Computer Science at UNamur. 

On a production line, images are captured by cameras positioned above a conveyor belt, and a computer system is then tasked with verifying, at a high rate, that each part meets specifications. “However, while the traditional inspection system does its job well, it also generates a lot of false positives—false alerts. This has a significant impact on the work of the people responsible for confirming or dismissing the alerts: they are constantly called upon, at the risk of wasting time and energy.” It is this process that Arnaud Bougaham decided to improve for his company by beginning a doctoral thesis in 2020 at the HuMaLearn laboratory in the School of Computer Science—a lab recognized for its expertise in machine learning—under the supervision of Professors Benoit Frenay and Isabelle Linden.

Today, the results of his dissertation were just presented, and they look promising. “I have developed a model that, using artificial intelligence, very accurately and reliably distinguishes a normal image from one that is not, thereby reducing the false alarms of the past.” 

Arnaud Bougaham

In addition, when a defect is suspected, the model can precisely pinpoint the affected area: the operator can view the indication, manipulate the part if necessary, and then make a decision. 

Another benefit of the tool? The system doesn’t just display a verdict; it also indicates the level of confidence the system has in its prediction. “The goal is to filter out simple cases and draw attention to more complex ones, so that experts are no longer overwhelmed by incorrect diagnoses and can focus on what really matters,” emphasizes Arnaud Bougaham.

Potential Beyond the Automotive Industry

While the tool was primarily designed to meet the needs of the automotive manufacturing industry, its underlying logic is generic. “It could be adapted for use in factories that manufacture hardwood flooring, wood, or fabric. But also in the medical sector,” says Arnaud Bougaham enthusiastically. “A disease is, ultimately, an anomaly in an image. The challenges are similar: when there’s a defect, it’s essential to detect it while minimizing false positives as much as possible. In the medical sector, having a similar tool that streamlines the analysis while leaving the final decision to the doctor can make a real difference.”

In this approach, AI is not an autopilot, but a decision-making tool that serves humans. “The goal is for the person to make the final decision, not for it to happen automatically,” adds Arnaud Bougaham. It is an assistant that “filters, explains, and provides all the information so that, ultimately, a person can make the decision.” 

Scan

When Research Meets “Real-World Truth”

Arnaud Bougaham’s dissertation grew out of a dual role: as a doctoral student at UNamur and, at the same time, as an engineer at AISIN Europe. This position gave him access to what he calls “the reality on the ground.” “In the lab, you can get very good results, but in production, there’s a reality with a specific environment.” Noisy data, variable conditions, cost constraints, and above all, production pace: “Our system has to operate at the same pace as the production line.” By dividing his time between AISIN Europe and UNamur, Arnaud Bougaham demonstrates how a dissertation can have a tangible impact on real-world needs—all while balancing the operations and obligations of both organizations. 

Image
Arnaud Bougaham

On the academic side, we tend to focus more on in-depth analysis to find long-term solutions, whereas on the industrial side, we’re more concerned with finding a reliable solution right here and now. This is where, when working simultaneously for both sectors, we must demonstrate a willingness to compromise, engage in dialogue, and build trust. But this synergy between academic research and industrial reality allows us to achieve very fruitful results.

Arnaud Bougaham Data scientist at AISIN EUROPE and research associate at the Faculty of Computer Science at UNamur

Roll out, expand, strengthen

Beyond the promises, this research is already being applied to real-world use cases. At AISIN Europe, the framework developed is currently being integrated into an active production line to detect unexpected components (such as a screw) on printed circuit boards. In the medical field, the approach is geared toward analyzing responsiveness in comatose patients through medical imaging that allows for the observation of organ and tissue function. Certain techniques are also being adapted for applications involving the segmentation of ovarian cancer or lymphoma.

These results also demonstrate how artificial intelligence can be reliable and human-centered in high-risk environments.

The Project in Pictures: AI Serving People in the Industrial Sector—Myth or Reality? :

IA

This article is from the "Impact" section of Omalius magazine, Issue #41 (June 2026).

Omalius 41

30 Years of Part-Time Education: When Adults Become Students, Too

Law
Economy
Management
Marketing
Computer science (IT)
Students

For more than 30 years, UNamur has offered flexible-schedule programs, allowing people already in the workforce to enroll in bachelor’s, master’s, and specialized master’s degree programs. These programs are based on a tailored teaching approach centered on the students’ experiences and real-world situations. 

Etudiante en reprise d'études travaillant sur son ordinateur portable

For as long as he can remember, Théophile Renier, 31, has always wanted to be a lawyer. It’s a calling that this freelance music industry professional—who is also a marine technician—set aside for a time... before returning to it a few years ago by enrolling in the School of Law. Since 2016, UNamur has indeed offered a bachelor’s and master’s program in law with a flexible schedule.  “What I find exceptional about this program is that the professors actively support students in their success,” says Théophile. “Everything is set up so that people who can’t be there in person can still access the courses remotely. ” All classes in the flexible-schedule law program are recorded and posted online to facilitate remote learning. Students are also notified of assignment and exam deadlines well in advance, to accommodate their professional and family commitments. 

As Mathieu Rolain, the program’s academic coordinator, points out, students in the flexible-schedule law program stand out for their exceptional determination and discipline: “superheroes,” he says. So, despite working three days a week at the Port of Zeebrugge and running his own music business—“I work about 50 hours a week”—Théophile manages to juggle it all. “For me, it’s actually very complementary,” says this passionate student. “People often wonder if it’s harder to study when you already have professional experience, but for me, studying is a pleasure! Nowadays, I focus more on what’s essential.” After completing his first two years of his bachelor’s degree, the student is now starting his third year; he then plans to pursue a master’s degree before joining the bar. “For me, it’s a childhood dream coming true,” he says. The future lawyer has just one piece of advice for those who are still hesitating: 

Many people wonder if it’s possible, or if they’re too old: in my opinion, you should put your reservations aside! This program is truly designed for people who are currently working: it has everything you need to succeed. 

Théophile Renier Law student

Soft Skills and Skills Assessment

For the past 30 years, the EMCP Faculty (Economics, Management, Communication, Sciences Po) has also offered flexible-schedule programs. It was in this context that the FSE+ track “Assessment, Strengthening, and Enhancement of Skills,” led by Jean-Yves Gnabo, was developed. Thanks to external support from the European Social Fund Plus, this educational program now provides better support for adults returning to school, not only at EMCP but also, eventually, in other faculties at UNamur. The project’s uniqueness lies in its integration of two complementary dimensions. On the one hand, an academic framework centered on an innovative thesis format, designed to prevent students from dropping out near the end of the program due to “writer’s block.” On the other hand, there is support for transferable skills: communication, organization, leadership, teamwork, time management, stress management, and more.

“Historically, we’ve focused on subject-specific skills, while soft skills were assumed to be acquired,” explains Jean-Yves Gnabo. “But succeeding in a part-time program also depends on ‘soft’ skills.” Furthermore, there was another blind spot in our approach: the fact that students were not always aware of their own skills. ” The FSE+ program therefore allows students, with the help of a coach, to conduct a skills assessment upon entry and upon completion, thereby measuring their progress. “This is not only essential for self-confidence but also for learning how to leverage newly acquired skills in the professional world,” explains the economics professor. For the past two years, the EMCP Faculty has also offered a specialized master’s program with a flexible schedule focused on management and sustainable economics. Thirty-five students were already enrolled in the first year. “Once the master’s program is established, the idea is to draw on a board of external stakeholders facing transition challenges, as well as to build a community among students, alumni, and faculty,” concludes Jean-Yves Gnabo. 

IT Work Shifts: Training That's More Necessary Than Ever

For the past 25 years, it has also been possible to earn a master’s degree in computer science at UNamur through a part-time program. While the sector faces many questions related to the widespread adoption of AI, Vincent Englebert, a professor of computer science, has no doubt that this program will continue to attract many working professionals. “We’re obviously going through a turbulent phase, but one thing is certain: while AI boosts the number of lines of code we can write per day, it also introduces a fair amount of errors into those lines, which creates serious maintenance issues... Companies are therefore beginning to realize the ‘technical debt’ generated by AI: if we ignore an engine problem just because a car is still running, we run the risk of the vehicle exploding one day on the highway... ” For Vincent Englebert, training new professionals in the challenges of generative AI is therefore a necessity. “I’d even say that AI is a great opportunity to pursue a degree in computer science today! Thirty years ago, computer scientists were asked to develop very basic programs. Today, they’re asked to develop much more complex systems, and AI—when used properly—frees them from highly technical tasks in this context. This allows them to focus on what really matters: what humans want to achieve with AI to make it more ethical, fairer, more useful, and safer.” 

Master 60 en sciences économiques à horaire décalé

A teaching approach tailored to students' needs

At UNamur, adult education plays a key role. Some of these working adults return to university to advance their careers, while others do so because they are in the process of changing careers. They all arrive with their own experiences, expectations, and dreams, but also with family and professional responsibilities, time constraints, and doubts… That is why, through its various flexible-schedule programs (evening and Saturday classes), UNamur has developed a specific pedagogical approach to best support these adults in their new academic and professional ambitions: realistic time management, an emphasis on collaborative work, close interaction with faculty, and a co-constructive approach to learning...

This article is taken from the "Guest" section of Omalius magazine, Issue #42 (September 2026).

Omalius-Septembre-2026

A study by VUB and UNamur shows that AI agents develop their own “artificial natural language” to communicate about complex data

Artificial Intelligences (AI)

AI systems spontaneously develop their own language, tailored to their sensors, rather than copying human language.

IA

Researchers from the VUB’s Artificial Intelligence Lab and the University of Namur (UNamur) demonstrate, in a new large-scale study, how autonomous AI systems can spontaneously develop, without human intervention, a functional and shared language. Unlike large language models such as ChatGPT, which learn existing human language by heart, these AI agents construct their own communication system, directly rooted in their specific sensors and environment. This discovery represents a major breakthrough for communication between autonomous systems, such as self-driving cars, robots and industrial sensors.

A layer of abstraction for non-human senses

Current AI language models focus on understanding and generating human languages such as Dutch or English. These human languages are historically rooted in our own sensory perceptions. When physical or virtual AI agents are equipped with sensors for which humans have no direct sensory equivalent – such as lidar, sonar, infrared or chemical analysis – existing human language proves insufficient.

Jérôme Botoko Ekila, a PhD student at the VUB’s AI Lab, explains: ‘Instead of forcing AI systems to communicate in a human language that does not correspond to their sensors, we allow them to develop their own “artificial natural language”. They construct effective concepts that correspond exactly to what they perceive via their sensors, even if the data is incomprehensible to humans.’

As, in practice, the sensors on different machines or vehicles can never be calibrated identically, the researchers believe that this language offers the ideal solution.

Paul Van Eecke, professor of artificial intelligence at the VUB: “In industrial or physical applications, it is simply unrealistic to expect all the sensors on different robots or vehicles to be calibrated in exactly the same way. Humans solve this problem through language: if I say I am standing near the ‘red car’, we understand each other without our eyes registering exactly the same colour values. Language provides a layer of abstraction above the sensor readings, and this is precisely what we are now allowing these AI agents to construct autonomously.”

Self-organisation and evolution

Instead of pre-programmed words or fixed codes, the researchers simply provided the agents with the basic mechanisms underpinning the evolution of human language: the local invention of words, the adaptation of conceptual prototypes, and the incorporation of feedback following successful or failed communication.

Image
Katrien Beuls

What’s fascinating is that there’s no central computer imposing the rules. Each time, we let two agents communicate with one another at a local level. It is solely through these individual interactions and updates that a global convention emerges spontaneously across the entire population. This principle of self-organisation allows the language to optimise itself automatically: the agents create exactly the number of categories needed to carry out their tasks efficiently, without superfluous details.

Katrien Beuls Professor in the School of Computer Science and a member of the NaDI Institute at UNamur

Tested on 37 datasets

The universal applicability of the method has been demonstrated across 37 varied and complex datasets. The agents successfully developed their own vocabulary to communicate, notably regarding physico-chemical analyses of wine, NASA data on exoplanets, visual datasets, and images of birds or faces. In all these situations, the agents successfully synthesised the data into new, shared concepts.

Transparent and adaptive

Transparency is a notable advantage of the systems developed. Unlike many ‘black-box’ AI models, researchers can precisely determine the meaning of a specifically developed word, as the prototype can be directly linked to the values of the underlying sensors.

Furthermore, these evolving linguistic systems are extremely robust. When the environment changes, when sensors produce noise, or when individual agents have a different structure, the language automatically adapts to the new circumstances through continuous interaction.

Reference

Botoko Ekila, J., Verheyen, L., Nevens, J., Beuls, K., & Van Eecke, P. (2026). Emergent Communication in Continuous Worlds: Self-Organization of Conceptually Grounded Vocabularies at Scale. Transactions of the Association for Computational Linguistics, vol. 14, pp. 2234–2257. DOI: 10.1162/TACL.a.799

Funding and collaboration

This research was carried out as part of a collaboration between the Artificial Intelligence Laboratory at the Vrije Universiteit Brussel (VUB) and the Faculty of Computer Science at the University of Namur (UNamur), and was funded by the Research Foundation – Flanders (FWO) and the Belgian National Research Fund (FNRS) as part of the WEAVE project.

FNRS-FWO

Research from UNamur Makes Artificial Intelligence Tools More Reliable

Artificial Intelligences (AI)
Computer science (IT)
Digital transition

Researchers at UNamur have developed a new method that has uncovered 32 previously undetected bugs in some of the most widely used software libraries for training artificial intelligence models. Originating from a master’s thesis, this research has already led to fixes in several software programs used by millions of developers around the world. 

Une recherche de l'UNamur rend les outils de l'intelligence artificielle plus fiables

Before artificial intelligence can recognize an image, understand a text, or answer a question, it must first be trained. To do this, developers and data scientists use specialized software libraries—true “toolboxes” that enable the construction of these AI models.

A research team comprising the University of Namur, the University of Passau (Germany), and TU Delft (Netherlands) has been studying this software. Their goal: to detect bugs that are particularly difficult to spot and that could cause a program to crash suddenly.

A New Method for Detecting Hidden Bugs

To verify that these libraries are functioning properly, researchers use software capable of automatically generating thousands of tests. The purpose of these tests is to push the programs to their limits in order to identify any potential errors. 

Until now, however, these tools had a major limitation: whenever a test caused a software library to malfunction, the software responsible for running the tests would also stop working. This made it impossible to pinpoint the exact source of the problem or report it to the developers. 

Researchers at UNamur have therefore devised a new method. Now, each test is run separately. If one of them causes a crash, it is isolated: the testing software continues to run, logs the error, and carries on with its analysis. This makes it possible to automatically detect bugs that previously went unnoticed. 

32 bugs discovered and already fixed

Using this new method, researchers analyzed 1,648 modules from the leading machine learning libraries and uncovered 32 previously unknown bugs. Several of these affected software that has become indispensable for training artificial intelligence models.

These issues were reported to the development teams. Several have already been acknowledged and fixed. “When a bug is fixed at this level, potentially thousands, or even millions, of users benefit,” notes Xavier Devroey, a professor in the Faculty of Computer Science at UNamur and a member of the PReCISE research center at the Namur Digital Institute (NaDI). Beyond these fixes, this research helps make the software on which so many artificial intelligence applications rely today more reliable.

Research that grew out of a thesis

This breakthrough originated from Lucas Berg’s master’s thesis, completed at UNamur under the supervision of Professor Xavier Devroey. As part of his research internship, Lucas Berg then joined Professor Annibale Panichella’s team at TU Delft to work on Pynguin, the open-source automated test generation tool developed at the University of Passau by Professor Gordon Fraser’s team, mentioned earlier. 

Initially, his goal was to improve this tool. By preventing crashes while discovering bugs, this first contribution subsequently caught the attention of researchers at the University of Passau, who were facing the same problem, leading to an international collaboration that would result in a joint scientific publication. 

Lucas Berg is currently pursuing a Ph.D. at the Faculty of Computer Science at UNamur under the supervision of Professors Wim Vanhoof and Xavier Devroey. The results of this research were presented at the IEEE International Conference on Software Testing, Verification, and Validation (ICST 2026), one of the most prestigious international conferences in the field of software testing.  

This research also received support from the Wallonia Public Service (SPW Research) and Wallonia-Brussels International. This recognition underscores the quality of this work and its significance to the international scientific community. 

The Namur Digital Institute (NaDI)

At NaDI, researchers develop innovative solutions to the new societal challenges posed by the digital revolution (eGov, eHealth, eServices, big data, etc.). Drawing from a variety of disciplines, the researchers combine their expertise in computer science, technology, ethics, law, management, and sociology. Comprising six research centers, the Namur Digital Institute offers unique multidisciplinary expertise in all areas of computer science, its applications, and its social impact.  

AI and Quality Control: When Humans Stay in Control

Artificial Intelligences (AI)
Computer science (IT)
Medicine

How can anomaly detection on a production line be improved? It was in response to a very concrete problem encountered at the company where he works that Arnaud Bougaham, with the support of his employer, decided to devote his thesis to this topic at the UNamur School of Computer Science. The goal? To develop an artificial intelligence model that assists operators in detecting anomalies on an industrial production line by reducing false alarms and making the diagnosis easier to understand. This approach also holds promise for deployment in the medical field. 

IA

Missing components, fragile solder joints, reversed polarity, scratches, or foreign objects such as a screw that falls onto the product: on a production line for automotive electronic components, such as the one at AISIN Europe, defects—even if they are rare—must be detected immediately and accurately. “My goal was to teach a computer to distinguish between what is normal and what is abnormal in images,” explains Arnaud Bougaham, a data scientist at AISIN Europe and an associate researcher at the Faculty of Computer Science at UNamur. 

On a production line, images are captured by cameras positioned above a conveyor belt, and a computer system is then tasked with verifying, at a high rate, that each part meets specifications. “However, while the traditional inspection system does its job well, it also generates a lot of false positives—false alerts. This has a significant impact on the work of the people responsible for confirming or dismissing the alerts: they are constantly called upon, at the risk of wasting time and energy.” It is this process that Arnaud Bougaham decided to improve for his company by beginning a doctoral thesis in 2020 at the HuMaLearn laboratory in the School of Computer Science—a lab recognized for its expertise in machine learning—under the supervision of Professors Benoit Frenay and Isabelle Linden.

Today, the results of his dissertation were just presented, and they look promising. “I have developed a model that, using artificial intelligence, very accurately and reliably distinguishes a normal image from one that is not, thereby reducing the false alarms of the past.” 

Arnaud Bougaham

In addition, when a defect is suspected, the model can precisely pinpoint the affected area: the operator can view the indication, manipulate the part if necessary, and then make a decision. 

Another benefit of the tool? The system doesn’t just display a verdict; it also indicates the level of confidence the system has in its prediction. “The goal is to filter out simple cases and draw attention to more complex ones, so that experts are no longer overwhelmed by incorrect diagnoses and can focus on what really matters,” emphasizes Arnaud Bougaham.

Potential Beyond the Automotive Industry

While the tool was primarily designed to meet the needs of the automotive manufacturing industry, its underlying logic is generic. “It could be adapted for use in factories that manufacture hardwood flooring, wood, or fabric. But also in the medical sector,” says Arnaud Bougaham enthusiastically. “A disease is, ultimately, an anomaly in an image. The challenges are similar: when there’s a defect, it’s essential to detect it while minimizing false positives as much as possible. In the medical sector, having a similar tool that streamlines the analysis while leaving the final decision to the doctor can make a real difference.”

In this approach, AI is not an autopilot, but a decision-making tool that serves humans. “The goal is for the person to make the final decision, not for it to happen automatically,” adds Arnaud Bougaham. It is an assistant that “filters, explains, and provides all the information so that, ultimately, a person can make the decision.” 

Scan

When Research Meets “Real-World Truth”

Arnaud Bougaham’s dissertation grew out of a dual role: as a doctoral student at UNamur and, at the same time, as an engineer at AISIN Europe. This position gave him access to what he calls “the reality on the ground.” “In the lab, you can get very good results, but in production, there’s a reality with a specific environment.” Noisy data, variable conditions, cost constraints, and above all, production pace: “Our system has to operate at the same pace as the production line.” By dividing his time between AISIN Europe and UNamur, Arnaud Bougaham demonstrates how a dissertation can have a tangible impact on real-world needs—all while balancing the operations and obligations of both organizations. 

Image
Arnaud Bougaham

On the academic side, we tend to focus more on in-depth analysis to find long-term solutions, whereas on the industrial side, we’re more concerned with finding a reliable solution right here and now. This is where, when working simultaneously for both sectors, we must demonstrate a willingness to compromise, engage in dialogue, and build trust. But this synergy between academic research and industrial reality allows us to achieve very fruitful results.

Arnaud Bougaham Data scientist at AISIN EUROPE and research associate at the Faculty of Computer Science at UNamur

Roll out, expand, strengthen

Beyond the promises, this research is already being applied to real-world use cases. At AISIN Europe, the framework developed is currently being integrated into an active production line to detect unexpected components (such as a screw) on printed circuit boards. In the medical field, the approach is geared toward analyzing responsiveness in comatose patients through medical imaging that allows for the observation of organ and tissue function. Certain techniques are also being adapted for applications involving the segmentation of ovarian cancer or lymphoma.

These results also demonstrate how artificial intelligence can be reliable and human-centered in high-risk environments.

The Project in Pictures: AI Serving People in the Industrial Sector—Myth or Reality? :

IA

This article is from the "Impact" section of Omalius magazine, Issue #41 (June 2026).

Omalius 41
All news

Agenda

  • 25
    2026
  • 27
    2026

CFIES 2026 - 9th International Symposium on the Teaching of Statistics

Congress / Colloquium / Conference

CFIES 2026 - 9th International Symposium on the Teaching of Statistics

Computer science
Didactics
25
2026 09:00 - 27
2026 17:00
Université de Namur - rue de Bruxelles, 61 - 5000 Namur

The International Francophone Symposium on the Teaching of Statistics will bring together, over the course of three days, educators and researchers from the relevant disciplines: statistics, of course, but also mathematics, educational sciences, knowledge engineering, the pedagogy of statistics, computer science, and, of course, disciplines that utilize statistical methods—whether in the exact sciences, the humanities, or the health sciences—as well as companies seeking to train their staff in statistical techniques.

Visule du colloque CFIES2026

One of the strengths of the conference is that it brings together stakeholders in lifelong learning.

The themes of this upcoming conference will address issues related to the teaching of statistics in the context of technical and technological innovations and transformations, from both historical and contemporary perspectives.

This conference is also being organized—for the first time—in collaboration with the “Socio-History of Statistics and Probability” group. 

26
2026

Public Defense of a Doctoral Dissertation in Computer Science - Maxime André

Thesis defense

Abstract

Over the past fifteen years, the microservices architectural style has become a dominant model, particularly for the design of cloud-based applications in the cloud continuum. Supported by agile development methods, it has been adopted by many actors of the modern software industry, including open-source communities as well as major software development companies such as Netflix, Spotify, and Amazon. Unlike the monolithic architectural style, it is praised for its qualities of modularity, heterogeneity, deployability, collaboration, and automation. Facilitating software evolution is one of the main motivations for adopting microservices architectures. Paradoxically, recent studies observe the opposite, especially concerning data manipulation layer. Nowadays, it is considered one of the most pressing challenges. Indeed, this style impacts the data layer, making it heterogeneous and distributed. In practice, developers encounter difficulties obtaining a holistic view of the system, understanding the related data concepts, measuring the impact of changes, and propagating subsequent modifications while preserving inter-microservices and inter-data relationships, which are sometimes implicit. These challenges often lead to manual effort during evolution tasks, for which developers suffer from a lack of support. This thesis aims to fill these gaps by proposing novel methods, techniques, and tools. More specifically, it combines static and dynamic program analysis techniques to perform reverse engineering of microservices and their data manipulation layer, to identify data concepts, and to locate related code fragments while taking into account their heterogeneous and distributed nature. It then visualizes the extracted information from a multi-layered viewpoint. Finally, it introduces techniques for analyzing the data-driven evolution of microservices from a historical perspective.

The members of the jury

  • Prof. Dr Katrien BEULS – President of the jury, UNamur, Belgium
  • Prof. Dr Anthony CLEVE – Supervisor, UNamur, Belgium
  • Prof. Dr Xavier DEVROEY – Internal member, UNamur, Belgium
  • Prof. Dr Etienne RIVIERE – Co-supervisor, UCLouvain, Belgium
  • Prof. Dr Massimiliano DI PENTA – External member, University of Sannio, Italy
  • Prof. Dr Michele LANZA – External member, Università della Svizzera Italiana, Switzerland
11
2027

Women in Science - 7th Edition

Event

Women in Science - 7th Edition

Gender and diversity
Science
SDG 5 - Gender equality
11
2027 09:00 - 20:00
Université de Namur, Faculté des sciences, auditoire S01 - rue Grafé, 2 - 5000 Namur
Contact person :  Comité Women in science

This annual event aims at promoting the access of women and girls to science and technology as well as their full and fair participation. It reminds the important role of women in the scientific community and constitutes a great opportunity to encourage girls and young women to participate in the scientific developments.

Women in science - 7th edition

Save the date!

Show your support for equal opportunities for ALL genders in science and technology.

More information to come. 

All events
Image
Etudiants en Fac d'informatique

Contacts and location

The Faculty of Computer Science welcomes you to the heart of the Namur campus.