Intrinsic dimensionality reduction with application to chemical simulation models

Simulation data from physical and chemical processes is typically high-dimensional. The dynamics of such systems can often be well described by a small number of coordinates. In this talk we present a method for nonlinear dimensionality reduction that embeds high-dimensional datasets into a low-dimensional intrinsic space. The constructed representation is utilized for merging two simulated datasets resulting from different partial observations of the same underlying dynamic process. Furthermore, we demonstrate how to extend the partially observed data in the ambient space from the reduced intrinsic coordinates.