Jesse Krijthe

As an assistant professor in the Pattern Recognition & Bioinformatics group of TU Delft (Delft University of Technology) I work on methodology and applications of statistical machine learning. Currently, I am particularly interested in causal inference, philosophy of statistics/data science, semi-supervised learning, model evaluation/selection, domain adaptation and missing data.

Previously I worked as a postdoc in the Data Science group of Radboud University Nijmegen on predictive and causal models for Parkinson’s disease.

During my PhD, obtained at Leiden University, I studied robust methods to do semi-supervised learning, that is: supervised models that can use additional unlabeled data with the property that, with high probability, performance is better than the original supervised model. During my PhD I was affiliated with the Pattern Recognition & Bioinformatics group of Delft University of Technology and the Department of Molecular Epidemiology of the Leiden University Medical Center.

You can find links to my R packages (Rtsne, RSSL) on GitHub, and further information on my website.

Publications