Projects
The Features of Fuzzy Representations: How do language learners specify foreign sounds? University of Antwerp
Iterative and multi-level methods for Bayesian multirelational factorization with features. KU Leuven
Many machine learning problems (classification, clustering, etc.) can be formulated as a factorization of an incompletely filled matrix where the goal is to predict the unknown values. These methods have been successful in large-scale recommender systems, like the Netflix challenge aimed at predicting movie preferences of users from 200 million movie ratings from half a million users. In particular, we are now generalizing the Bayesian ...
Device-free crowd sensing at large music festivals using radio frequency signal features. University of Antwerp
RAPTOR: a novel multi-view integrative framework to identify key features of T cell based immunology. University of Antwerp
A quest for the perfect brand name: the role of physical brand name features on brand memory, preference and performance Vrije Universiteit Brussel
importance in establishing good brand names. Distinctive brand
names are overall better remembered and for the category of luxury
brands, they have been demonstrated to be preferred over non- or
less distinctive brand names. Although studies have primarily
focused on semantical brand name features to operationalize the
concept of ...
Postpartum breast cancer diagnosed during involution: a distinct entity with unique clinicopathological, molecular and immunological features? KU Leuven
Around 10-25% of breast cancers in young women are diagnosed during pregnancy or the postpartum years. Given the trend to delay childbearing, this frequency is expected to increase. Emerging evidence from our group and others indicates that breast cancers diagnosed in the postpartum period (PPBC) have a worse prognosis, with an increased risk for metastasis and death than breast cancers occurring during or outside pregnancy. The ...
Logical Relations and Secure Compilation for Advanced Programming Language Features using Iris KU Leuven
Compilers translate programs written in a high-level programming language (PL) to the low-level machine language which can be executed by the computer hardware. Modern PLs use types, to detect errors in programs, e.g., subtracting a string of characters (a term of string type) from an integer (a term of numeral type). Advanced PLs, e.g., ML, Rust, etc., are usually designed so that well-typed programs have certain desired properties. These ...