Parametric nonlinear dimensionality reduction using kernel t-SNE Andrej Gisbrecht Alexander Schulz Barbara Hammer University of Bielefeld - CITEC centre of excellence, Germany Preprint of the publication [10], as provided by the authors. Abstract Novel non-parametric dimensionality reduction techniques such as t-distributed stochastic neighbor embedding (t-SNE) lead to a powerful and flexible visualization of high-dimensional data. One drawback of non-parametric techniques is their lack of an explicit out-of-sample extension. In this contribution, we propose an efficient extension of t-SNE to a parametric framework, kernel t-SNE, which preserves the flexibility of basic t-SNE, but enables explicit out-of-sample extensions. We test the ability of kernel t-SNE in comparison to standard t-SNE for benchmark data sets, in particular addressing the generalization ability of the mapping for novel data. In the context of large data sets, this procedure enables us to train a mapping for a fixed size subset only, mapping all data afterwards in linear time. We demonstrate that this technique yields satisfactory results also for large data sets provided missing information due to the small size of the subset is accounted for by auxiliary information such as class labels, which can be integrated into kernel t-SNE based on the Fisher information. 1 Introduction Handling big data constitutes one of the main challenges of information technology in the new century, incorporating, among other issues, the task to create ‘effective human-computer interaction tools for facilitating rapidly customizable visual reasoning for diverse missions’ [13]. In this context, the visual inspection of high-dimensional data sets offers an intuitive interface for humans to rapidly detect structural elements of the data such as clus- ters, homogeneous regions, or outliers, relying on the astonishing cognitive 1