Density peak clustering using tensor network

Xiao Shi, Yun Shang

Abstract: We introduce a density-based clustering algorithm with tensor networks. In order to demonstrate its effectiveness, we apply it to various types of data sets, including synthetic data sets, real world data sets, and computer vision data sets. Results demonstrate that it is an efficient quantum-inspired unsupervised learning algorithm and can recognize clusters of arbitrary shape and size. It can also be seen that large quantum entanglement tends to provide better clustering results.