References & Citations
Quantum Physics
Title: Quantum Spectral Clustering
(Submitted on 1 Jul 2020 (this version), latest version 14 Jun 2021 (v4))
Abstract: Spectral clustering is a powerful unsupervised machine learning algorithm for clustering data with non convex or nested structures. With roots in graph theory, it uses the spectral properties of the Laplacian matrix to project the data in a low dimensional space where clustering is more efficient. Despite its success in clustering tasks, spectral clustering suffers in practice from a fast-growing running time of $O(n^3)$, where $n$ is the number of points in the dataset. In this work, we propose the first algorithm performing spectral clustering on a quantum computer with provable guarantees, extending a number of works in quantum machine learning. The quantum algorithm is composed of two parts: the first is the efficient creation of the quantum state corresponding to the projected Laplacian matrix, and the second consists in applying the existing quantum analogue of the $k$-means algorithm. Both steps depend polynomially on the number of clusters, as well as precision and data parameters arising from quantum procedures, and polylogarithmically on the dimension of the input vectors. Our numerical simulations show an asymptotic linear growth with $n$ when all terms are taken into account, considerably better than the classical cubic growth. This work opens the path to other graph based quantum machine learning algorithms, as it provides the first routines for efficient computation and quantum access to the Incidence, Adjacency, and projected Laplacian matrices of a graph.
Submission history
From: Jonas Landman [view email][v1] Wed, 1 Jul 2020 07:11:42 GMT (663kb,D)
[v2] Thu, 18 Mar 2021 15:19:08 GMT (1442kb,D)
[v3] Tue, 27 Apr 2021 12:28:59 GMT (1237kb,D)
[v4] Mon, 14 Jun 2021 07:16:18 GMT (1444kb,D)
Link back to: arXiv, form interface, contact.