معرفی
Rebecca Marion is a Researcher at UCLouvain specializing in machine learning, with expertise in dimensionality reduction and feature selection. She holds a Doctor of Science from UCLouvain (awarded 2021) for her work on statistical and machine learning methods for identifying variable clusters.
Her education:
- Doctor of Science, Statistical and Machine Learning Methods for Identifying Clusters of Variables, UCLouvain (2021)
Marion's research focuses on nonlinear dimensionality reduction (e.g., t-SNE, UMAP) and feature selection stability. She develops gradient-based explanation methods for complex embeddings and investigates the robustness of machine learning models. Her work bridges theoretical advances with practical data science applications.
Recent publications highlight her contributions to explainable AI: a 2024 paper on gradient-based explanations for nonlinear embeddings and a 2023 study improving feature selection stability in regression. These works address critical challenges in interpretability and reliability of machine learning systems.
No scientific awards are listed in the available profile information.
Information about student advising and research grants is not provided in the current data.
She is part of UCLouvain's machine learning research community, though specific lab or team affiliations are not detailed.
