Jul 12, 2022 · We propose a refactoring of this algorithm by posing its computation as matrix operations, specifically Khatri-Rao products and standard matrix multiplications.

Mar 1, 2022 · Higher order moment tensors contains important statistical information when the data is non-Gaussian. For example, anomaly detection with ...

ABSTRACT. The decomposition of higher-order joint cumulant tensors of spatio- temporal data sets is useful in analyzing multi-variate non-Gaussian.

Li, Zitong, Kolla, Hemanth, and Phipps, Eric. 2022. "Parallel Memory-Efficient Computation of Symmetric Higher-Order Joint Moment Tensors.". United States.

Duration: 1:06
Posted: Jun 14, 2022

ABSTRACT. The decomposition of higher-order joint cumulant tensors of spatio- temporal data sets is useful in analyzing multi-variate non-Gaussian.

We propose a refactoring of this algorithm by posing its computation as matrix operations, specifically Khatri-Rao products and standard matrix multiplications.

In this paper, we introduce a novel algorithm for calculating arbitrary order cumulants of multidimensional data.

The algorithm provided in the paper takes advantage of super- symmetry of cumulant and moment tensors. We show that the proposed algorithm considerably reduces ...

Parallel memory-efficient computation of symmetric higher-order joint moment tensors. Z Li, H Kolla, ET Phipps. Proceedings of the Platform for Advanced ...