Improving data analysis in biology and in general with Tensor Decomposition

This talk will provide an introduction to the basic principles of tensor decomposition methods, especially CANDECOMP/PARAFAC (CP) decomposition. I will explain how such methods can be used to extract meaningful and interpretable patterns from high-dimensional tensor-structured data, which commonly arises in biology, as well as in a broad range of other scientific domains. These patterns can then be used to gain a better understanding of the phenomena under study. I will also briefly discuss how tensor decomposition methods can be extended for various types of data, focusing in particular on how I have been trying to better model longitudinal data.