Skolkovo Institute of Science and Technology, Moscow
and Artificial Intelligence project, Riken
Plenary lecture
Machine learning and tensor networks and their applications, in brain computer interface, neurofeedback and recognition of human emotions
Tensor decomposition (TD) and their generalizations tensor networks (TNs) are promising, and emerging tools in Machine Learning (ML), Big Data Analysis (BDA) and Deep Learning (DL).Many real-life data can be naturally described as higher-order tensors which can be represented in distributed and compressed forms by low-rank factorization, with substantially reduced the number of parameters. We will present a brief overview of tensor decomposition in components or factors and tensor networks architectures and associated learning algorithms and indicate their perspective and potential applications. Special emphasis will be given to feature extraction,classification, clustering and anomaly detection problems in computational neuro-science, especially brain computer interface. We graphically illustrate models of Tensor Train, Tensor Ring and Hierarchic Tucker and other promising tensor network decomposition for high order tensors.
