Parallel Computation of Multi-Slice Clustering of Third-Order Tensors

TL;DR

This paper introduces parallel algorithms for Multi-Slice Clustering of third-order tensors, enabling scalable spectral analysis on large datasets using distributed memory systems.

Contribution

It presents a novel parallel scheme for MSC that improves performance and scalability over sequential methods for third-order tensor clustering.

Findings

Parallel MSC outperforms sequential algorithms.

The method scales effectively on distributed memory systems.

Spectral analysis enables efficient clustering of large tensor datasets.

Abstract

Machine Learning approaches like clustering methods deal with massive datasets that present an increasing challenge. We devise parallel algorithms to compute the Multi-Slice Clustering (MSC) for 3rd-order tensors. The MSC method is based on spectral analysis of the tensor slices and works independently on each tensor mode. Such features fit well in the parallel paradigm via a distributed memory system. We show that our parallel scheme outperforms sequential computing and allows for the scalability of the MSC method.