A parallel multi‐block alternating direction method of multipliers for tensor completion
Abstract This paper proposes an algorithm for the tensor completion problem of estimating multi‐linear data under the limitation of observation rate. Many tensor completion methods are based on nuclear norm minimization, they may fail to achieve the global solution for solving nuclear norm minimizat...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2021-11-01
|
| Series: | IET Image Processing |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/ipr2.12289 |
| _version_ | 1797901434234077184 |
|---|---|
| author | Hu Zhu Zhongyang Wang Taiyu Yan Yu‐Feng Yu Lizhen Deng Bing‐Kun Bao |
| author_facet | Hu Zhu Zhongyang Wang Taiyu Yan Yu‐Feng Yu Lizhen Deng Bing‐Kun Bao |
| author_sort | Hu Zhu |
| collection | DOAJ |
| description | Abstract This paper proposes an algorithm for the tensor completion problem of estimating multi‐linear data under the limitation of observation rate. Many tensor completion methods are based on nuclear norm minimization, they may fail to achieve the global solution for solving nuclear norm minimization in tensor completion problem with high missing ratio. To tackle this issue, an adaptive tensor completion method based on parallel multi‐block alternating direction method of multipliers (ADMM) algorithm is proposed, it can derive the model from the initial estimate and compute the next estimate from the current solution. The parallel multi‐block ADMM with global convergence is adopted to solve the dual problem, which greatly improves the processing power and reliability of the algorithm. |
| first_indexed | 2024-04-10T09:02:01Z |
| format | Article |
| id | doaj.art-3ba155097d234187b936e29be16ac7da |
| institution | Directory Open Access Journal |
| issn | 1751-9659 1751-9667 |
| language | English |
| last_indexed | 2024-04-10T09:02:01Z |
| publishDate | 2021-11-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Image Processing |
| spelling | doaj.art-3ba155097d234187b936e29be16ac7da2023-02-21T11:57:05ZengWileyIET Image Processing1751-96591751-96672021-11-0115133053306210.1049/ipr2.12289A parallel multi‐block alternating direction method of multipliers for tensor completionHu Zhu0Zhongyang Wang1Taiyu Yan2Yu‐Feng Yu3Lizhen Deng4Bing‐Kun Bao5National Engineering Research Center of Communication and Network Technology Nanjing University of Posts and Telecommunications Nanjing ChinaNational Engineering Research Center of Communication and Network Technology Nanjing University of Posts and Telecommunications Nanjing ChinaSchool of Life Science and Technology University of Electronic Science and Technology of China Chengdu ChinaDepartment of Statistics Guangzhou University Guangzhou ChinaJiangsu Province Key Lab on Image Processing and Image Communication Nanjing University of Posts and Telecommunications Nanjing ChinaCollege of Telecommunications and Information Engineering Nanjing University of Posts and Telecommunications Nanjing ChinaAbstract This paper proposes an algorithm for the tensor completion problem of estimating multi‐linear data under the limitation of observation rate. Many tensor completion methods are based on nuclear norm minimization, they may fail to achieve the global solution for solving nuclear norm minimization in tensor completion problem with high missing ratio. To tackle this issue, an adaptive tensor completion method based on parallel multi‐block alternating direction method of multipliers (ADMM) algorithm is proposed, it can derive the model from the initial estimate and compute the next estimate from the current solution. The parallel multi‐block ADMM with global convergence is adopted to solve the dual problem, which greatly improves the processing power and reliability of the algorithm.https://doi.org/10.1049/ipr2.12289Optimisation techniquesInterpolation and function approximation (numerical analysis)Optimisation techniques |
| spellingShingle | Hu Zhu Zhongyang Wang Taiyu Yan Yu‐Feng Yu Lizhen Deng Bing‐Kun Bao A parallel multi‐block alternating direction method of multipliers for tensor completion IET Image Processing Optimisation techniques Interpolation and function approximation (numerical analysis) Optimisation techniques |
| title | A parallel multi‐block alternating direction method of multipliers for tensor completion |
| title_full | A parallel multi‐block alternating direction method of multipliers for tensor completion |
| title_fullStr | A parallel multi‐block alternating direction method of multipliers for tensor completion |
| title_full_unstemmed | A parallel multi‐block alternating direction method of multipliers for tensor completion |
| title_short | A parallel multi‐block alternating direction method of multipliers for tensor completion |
| title_sort | parallel multi block alternating direction method of multipliers for tensor completion |
| topic | Optimisation techniques Interpolation and function approximation (numerical analysis) Optimisation techniques |
| url | https://doi.org/10.1049/ipr2.12289 |
| work_keys_str_mv | AT huzhu aparallelmultiblockalternatingdirectionmethodofmultipliersfortensorcompletion AT zhongyangwang aparallelmultiblockalternatingdirectionmethodofmultipliersfortensorcompletion AT taiyuyan aparallelmultiblockalternatingdirectionmethodofmultipliersfortensorcompletion AT yufengyu aparallelmultiblockalternatingdirectionmethodofmultipliersfortensorcompletion AT lizhendeng aparallelmultiblockalternatingdirectionmethodofmultipliersfortensorcompletion AT bingkunbao aparallelmultiblockalternatingdirectionmethodofmultipliersfortensorcompletion AT huzhu parallelmultiblockalternatingdirectionmethodofmultipliersfortensorcompletion AT zhongyangwang parallelmultiblockalternatingdirectionmethodofmultipliersfortensorcompletion AT taiyuyan parallelmultiblockalternatingdirectionmethodofmultipliersfortensorcompletion AT yufengyu parallelmultiblockalternatingdirectionmethodofmultipliersfortensorcompletion AT lizhendeng parallelmultiblockalternatingdirectionmethodofmultipliersfortensorcompletion AT bingkunbao parallelmultiblockalternatingdirectionmethodofmultipliersfortensorcompletion |