Abstract
Tensors are increasingly ubiquitous in various areas of applied mathematics and computing, and tensor decompositions are of practical significance and benefit many applications in data completion, image processing, computer vision, collaborative filtering, etc. Recently, Kilmer and Martin propose a new tensor factorization strategy, tensor singular value decomposition (t-svd), which extends the matrix singular value decomposition to tensors. However, computing t-svd for high dimensional tensors costs much computation and thus cannot efficiently handle large scale datasets. Motivated by advantage of quantum computation, in this paper, we present a quantum algorithm of t-svd for third-order tensors and then extend it to order-p tensors. We prove that our quantum t-svd algorithm for a third-order N dimensional tensor runs in time O(Npolylog(N)) if we do not recover classical information from the quantum output state. Moreover, we apply our quantum t-svd algorithm to context-aware multidimensional recommendation systems, where we just need to extract partial classical information from the quantum output state, thus achieving low time complexity.
| Original language | English |
|---|---|
| Article number | 075001 |
| Pages (from-to) | 1-15 |
| Number of pages | 15 |
| Journal | Journal of Physics Communications |
| Volume | 5 |
| Issue number | 7 |
| DOIs | |
| Publication status | Published - 2 Jul 2021 |
Keywords
- Quantum fourier transform
- Quantum singular value estimation
- T-product
- T-svd
- Tensor singular value decomposition
ASJC Scopus subject areas
- General Physics and Astronomy
Fingerprint
Dive into the research topics of 'Quantum tensor singular value decomposition*'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver