Nonnegative tensor decomposition with custom clustering for microphase separation of block copolymers
B.S. Alexandrov, V. Stanev, V.V. Vesselinov, K. Rasmussen
Statistical Analysis and Data Mining2019DOI 10.1002/sam.11407
Summary
The emerging collections of large and distinct high-dimensional datasets increase the interest in factor analysis based on tensor decomposition [1]. These collected datasets include only directly observable quantities, while the underlying processes are either too complex and cannot be observed directly or are completely unknown. These processes are called latent variables or latent features [2].