Projects
Discovering biased representations of people KU Leuven
Standard approaches to detecting and mitigating bias assume a fixed representation of people and of decisions concerning them in data, with some fixed utility notion (e.g. accuracy) and some formal measure of bias/fairness. Project KULEUVEN-1 (Discovering biased representations of people) will focus on representational adequacy, i.e., the extent to which data represent what is (legally and/or ethically) objectionable about bias. The project ...
Identifying and Amplifying Non-Technical Skills in Software Engineering Education KU Leuven
Industrial software engineering is a young and complex field that is still riddled with challenges to tackle in order to successfully develop a product. These challenges are not limited to technical problems. Most students and applicants are able to quickly adapt to certain technical tools and programming languages but lack other essential abilities required in the field.
Imperfect technical skills for the aspiring software engineer are ...
Decision Analytics: complex decision processes KU Leuven
Enterprises often experience difficulties coping with complex decision-making processes. The new Decision Modeling Notation (DMN) standard tries to improve this by also allowing business users without ICT background to model decisions and decision making processes. This allows them to analyze them better, and to automate them (partly). In practice, this standard isn't currently being used very often in Flanders. The overall goal of this ...
Efficient and Versatile Methods for Relational Machine Learning KU Leuven
The field of machine learning concerns computer algorithms that automatically improve their performance on a task through experience from data. Most common machine learning approaches expect data in the form of a table. Therefore, they cannot directly learn from more expressive data formats, such as relational databases, knowledge bases or logic programs, which are able to represent relations between complex instances. To learn from these ...
Elevating Multithreading Further into the Cloud KU Leuven
Due to the surging popularity of cloud computing on one hand and the emergence of numerous novel, innately parallel workloads on the other, executing multithreaded applications in a virtualized setting has become common practice in industry. However, multithreading is known to be highly susceptible to severe performance degradation in virtualized environments. In response, virtualization technologies have evolved rapidly over the years; to ...
Applications of probabilistic logic KU Leuven
Applications (e.g. in the domain of bio-informatics) of probabilistic logic are investigated.
Analysis and transformation of programs. KU Leuven
The work is situated in formal methods and automatic techniques for the analysis and transformation of programs, specifically in the context of Logical Programming.
Formal theories for analyzing programs have been developed, especially in the context of Termination and Non-Termination of Programs. Numerous results were achieved that shifted existing boundaries around the undecidable “Halting” problem. This work was situated in the ...
Probabilistic Logic Learning and Computational Creativity KU Leuven
In the rapidly evolving domain of artificial intelligence, large language models have shown significant progress in generating text for creative applications, including humor. However, they lack the controllability, efficiency and explainability which symbolic AI methods offer. In this thesis, we explore how symbolic and subsymbolic AI methods can be used, extended and integrated for applications in creative AI, particularly for Dutch humor, ...