Sparse signal recovery via minimax‐concave penalty and ℓ1 ‐norm loss function
In sparse signal recovery, to overcome the ℓ1 ‐norm sparse regularisation's disadvantages tendency of uniformly penalise the signal amplitude and underestimate the high‐amplitude components, a new algorithm based on a non‐convex minimax‐concave penalty is proposed, which can approximate the ℓ0...
Uloženo v:
| Vydáno v: | IET signal processing Ročník 12; číslo 9; s. 1091 - 1098 |
|---|---|
| Hlavní autoři: | , , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
The Institution of Engineering and Technology
01.12.2018
|
| Témata: | |
| ISSN: | 1751-9683, 1751-9683 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Abstract | In sparse signal recovery, to overcome the ℓ1 ‐norm sparse regularisation's disadvantages tendency of uniformly penalise the signal amplitude and underestimate the high‐amplitude components, a new algorithm based on a non‐convex minimax‐concave penalty is proposed, which can approximate the ℓ0 ‐norm more accurately. Moreover, the authors employ the ℓ1 ‐norm loss function instead of the ℓ2 ‐norm for the residual error, as the ℓ1 ‐loss is less sensitive to the outliers in the measurements. To rise to the challenges introduced by the non‐convex non‐smooth problem, they first employ a smoothed strategy to approximate the ℓ1 ‐norm loss function, and then use the difference‐of‐convex algorithm framework to solve the non‐convex problem. They also show that any cluster point of the sequence generated by the proposed algorithm converges to a stationary point. The simulation result demonstrates the authors’ conclusions and indicates that the algorithm proposed in this study can obviously improve the reconstruction quality. |
|---|---|
| AbstractList | In sparse signal recovery, to overcome the ℓ1 ‐norm sparse regularisation's disadvantages tendency of uniformly penalise the signal amplitude and underestimate the high‐amplitude components, a new algorithm based on a non‐convex minimax‐concave penalty is proposed, which can approximate the ℓ0 ‐norm more accurately. Moreover, the authors employ the ℓ1 ‐norm loss function instead of the ℓ2 ‐norm for the residual error, as the ℓ1 ‐loss is less sensitive to the outliers in the measurements. To rise to the challenges introduced by the non‐convex non‐smooth problem, they first employ a smoothed strategy to approximate the ℓ1 ‐norm loss function, and then use the difference‐of‐convex algorithm framework to solve the non‐convex problem. They also show that any cluster point of the sequence generated by the proposed algorithm converges to a stationary point. The simulation result demonstrates the authors’ conclusions and indicates that the algorithm proposed in this study can obviously improve the reconstruction quality. |
| Author | Sun, Yuli Chen, Hao Tao, Jinxu |
| Author_xml | – sequence: 1 givenname: Yuli surname: Sun fullname: Sun, Yuli organization: University of Science and Technology of China – sequence: 2 givenname: Hao surname: Chen fullname: Chen, Hao organization: University of Science and Technology of China – sequence: 3 givenname: Jinxu surname: Tao fullname: Tao, Jinxu email: tjingx@ustc.edu.cn organization: University of Science and Technology of China |
| BookMark | eNpN0EtOwzAQBmALFYm2cAB2vkCK7fiRsIOKQqVKLApryzgTZJQ4kR0K2bHgAEjcsCfBFSxYzWhmNPr1zdDEdx4QOqdkQQkvLxwMWezDghFaLATNyRGaUiVoVsoin_zrT9AsxhdChBSUTdF225sQAUf37E2DA9huB2HEO2dw67xrzfv-48t23pod4B7S0TBi4yu8__ymOO18F1rcdDHi-tXbwXX-FB3Xpolw9lfn6HF187C8yzb3t-vl1SaLlNGUxhhmJTGFLaRUCkA8WQaM11YJVtZSQsmlEpxTZSuouFWMy7o0la3SjJF8ji5__765Bkbdh5Q2jJoSfSDRiUQnEn0g0QcSvV1v2PWKEFnQ_Ad-M16E |
| ContentType | Journal Article |
| Copyright | 2021 The Institution of Engineering and Technology |
| Copyright_xml | – notice: 2021 The Institution of Engineering and Technology |
| DOI | 10.1049/iet-spr.2018.5130 |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering |
| EISSN | 1751-9683 |
| EndPage | 1098 |
| ExternalDocumentID | SIL2BF00681 |
| Genre | article |
| GrantInformation_xml | – fundername: National Natural Science Foundation of China funderid: 81527802 |
| GroupedDBID | .DC 0R~ 0ZK 1OC 24P 29I 4.4 5GY 6IK 8FE 8FG AAHHS AAHJG AAJGR ABJCF ABMDY ABQXS ACCFJ ACCMX ACESK ACGFO ACGFS ACIWK ACXQS ADEYR ADZOD AEEZP AEGXH AENEX AEQDE AFKRA AIAGR AIWBW AJBDE ALMA_UNASSIGNED_HOLDINGS ALUQN ARAPS AVUZU BENPR BGLVJ CCPQU CS3 DU5 EBS EJD GROUPED_DOAJ HCIFZ HZ~ IAO IFIPE IGS IPLJI ITC J9A JAVBF K6V K7- L6V LAI LXU M43 M7S MCNEO NADUK NXXTH O9- OCL OK1 P2P P62 PTHSS RIE RNS RUI S0W UNMZH ~ZZ |
| ID | FETCH-LOGICAL-s1211-9aa2c60a8c86677ee5bc2e24fc7529f66e946754417cded4c7246f9adcd441203 |
| IEDL.DBID | 24P |
| ISSN | 1751-9683 |
| IngestDate | Wed Jan 22 16:31:07 EST 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 9 |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-s1211-9aa2c60a8c86677ee5bc2e24fc7529f66e946754417cded4c7246f9adcd441203 |
| PageCount | 8 |
| ParticipantIDs | wiley_primary_10_1049_iet_spr_2018_5130_SIL2BF00681 |
| PublicationCentury | 2000 |
| PublicationDate | December 2018 |
| PublicationDateYYYYMMDD | 2018-12-01 |
| PublicationDate_xml | – month: 12 year: 2018 text: December 2018 |
| PublicationDecade | 2010 |
| PublicationTitle | IET signal processing |
| PublicationYear | 2018 |
| Publisher | The Institution of Engineering and Technology |
| Publisher_xml | – name: The Institution of Engineering and Technology |
| References | 2010; 11 2015; 37 2006; 52 2017; 3 2012 2013; 22 2005; 133 2015; 244 2017; 65 2008; 14 2011; 33 2014; 24 2013; 140 2014; 61 2015; 8 2014; 23 2012; 52 2018; 69 2007; 16 2013; 19 2017; 74 2013; 51 2015; 64 2017 2016 1981 2013 2012; 6 2009; 2 2012; 23 2010; 4 2016; 26 2012; 21 |
| References_xml | – volume: 16 start-page: 1101 issue: 4 year: 2007 end-page: 1111 article-title: Deblurring of color images corrupted by impulsive noise publication-title: IEEE Trans. Image Process. – volume: 244 start-page: 26 issue: 1 year: 2015 end-page: 46 article-title: DC approximation approaches for sparse optimization publication-title: Eur. J. Oper. Res. – year: 1981 – volume: 37 start-page: 536 issue: 1 year: 2015 end-page: 563 article-title: Minimization of L1‐2 for compressed sensing publication-title: SIAM J. Sci. Comput. – volume: 23 start-page: 1013 issue: 7 year: 2012 end-page: 1027 article-title: L1/2 regularization: a thresholding representation theory and a fast solver publication-title: IEEE Trans. Neural Netw. Learn. Syst. – volume: 2 start-page: 323 issue: 2 year: 2009 end-page: 343 article-title: The split Bregman method for L1‐regularized problems publication-title: SIAM J. Imaging Sci. – volume: 65 start-page: 2561 issue: 10 year: 2017 end-page: 2575 article-title: Sparse signal approximation via non‐separable regularization publication-title: IEEE Trans. Signal Process. – volume: 52 start-page: 489 issue: 2 year: 2006 end-page: 509 article-title: Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information publication-title: IEEE Trans. Inf. Theory – volume: 23 start-page: 078703 issue: 7 year: 2014 article-title: Image reconstruction from few views by ℓ0‐norm optimization publication-title: Chin. Phys. B – volume: 19 start-page: 2277 issue: 5B year: 2013 end-page: 2293 article-title: Multi‐stage convex relaxation for feature selection publication-title: Bernoulli. (Andover) – volume: 51 start-page: 927 issue: 2 year: 2013 end-page: 957 article-title: Improved iteratively reweighted least squares for unconstrained smoothed Lq minimization publication-title: SIAM J. Numer. Anal. – volume: 61 start-page: 198 issue: 3 year: 2014 end-page: 202 article-title: New improved algorithms for compressive sensing based on LP norm publication-title: IEEE Trans. Circuits Syst. II, Express Briefs – volume: 69 start-page: 297 issue: 2 year: 2018 end-page: 324 – start-page: 37 year: 2013 end-page: 45 article-title: A general iterative shrinkage and thresholding algorithm for non‐convex regularized optimization problems – volume: 11 start-page: 1081 year: 2010 end-page: 1107 article-title: Analysis of multi‐stage convex relaxation for sparse regularization publication-title: J. Mach. Learn. Res. – volume: 65 start-page: 105 issue: 1 year: 2017 end-page: 118 article-title: Robust sparse recovery in impulsive noise via LP‐L1 optimization publication-title: IEEE Trans. Signal Process. – start-page: 4643 year: 2016 end-page: 4647 article-title: Robust sparse recovery for compressive sensing in impulsive noise using ℓp‐norm model fitting – volume: 52 start-page: 509 issue: 3 year: 2012 end-page: 535 article-title: Exact penalty and error bounds in DC programming publication-title: J. Glob. Optim. – volume: 65 start-page: 4481 issue: 17 year: 2017 end-page: 4494 article-title: Sparse regularization via convex analysis publication-title: IEEE Trans. Signal Process. – start-page: 1027 year: 2013 end-page: 1033 article-title: A concave conjugate approach for nonconvex penalized regression with the MCP penalty – start-page: 427 year: 2013 end-page: 435 article-title: Revisiting Frank‐Wolfe: projection‐free sparse convex optimization – volume: 26 start-page: 43 issue: 1 year: 2016 end-page: 54 article-title: MR imaging reconstruction using a modified descent‐type alternating direction method publication-title: Int. J. Imaging Syst. Technol. – volume: 4 start-page: 244 issue: 2 year: 2010 end-page: 254 article-title: Compressed synthetic aperture radar publication-title: IEEE. J. Sel. Top. Signal Process. – volume: 21 start-page: 397 issue: 1 year: 2012 end-page: 405 article-title: Improved image recovery from compressed data contaminated with impulsive noise publication-title: IEEE Trans. Image Process. – volume: 2 start-page: 183 issue: 1 year: 2009 end-page: 202 article-title: A fast iterative shrinkage‐thresholding algorithm for linear inverse problems publication-title: SIAM J. Imaging Sci. – volume: 14 start-page: 877 issue: 5–6 year: 2008 end-page: 905 article-title: Enhancing sparsity by reweighted L1 minimization publication-title: J. Fourier Anal. Appl. – volume: 140 start-page: 125 issue: 1 year: 2013 end-page: 161 article-title: Gradient methods for minimizing composite functions publication-title: Math. Program. – volume: 3 start-page: 566 issue: 4 year: 2017 end-page: 579 article-title: Efficient and robust recovery of sparse signal and image using generalized nonconvex regularization publication-title: IEEE Trans. Comput. Imaging – volume: 24 start-page: 215 issue: 3 year: 2014 end-page: 223 article-title: Few views image reconstruction using alternating direction method via 0norm minimization publication-title: Int. J. Imaging Syst. Technol. – volume: 6 start-page: 976 issue: 7 year: 2012 end-page: 985 article-title: Improved decision‐based detail‐preserving variational method for removal of random‐valued impulse noise publication-title: IET Image Process. – volume: 133 start-page: 23 issue: 1‐4 year: 2005 end-page: 46 article-title: The DC (difference of convex functions) programming and DCA revisited with DC models of real world nonconvex optimization problems publication-title: Ann. Oper. Res. – start-page: 895 year: 2012 end-page: 903 article-title: Robust multi‐task feature learning – year: 2017 – volume: 22 start-page: 4724 issue: 12 year: 2013 end-page: 4737 article-title: Efficient algorithms for robust recovery of images from compressed data publication-title: IEEE Trans. Image Process. – volume: 74 start-page: 1 issue: 2 year: 2017 end-page: 19 article-title: Fast L1–L2 minimization via a proximal operator publication-title: J. Sci. Comput. – volume: 8 start-page: 1798 issue: 3 year: 2015 end-page: 1823 article-title: A weighted difference of anisotropic and isotropic total variation model for image processing publication-title: SIAM J. Imaging Sci. – volume: 64 start-page: 178 issue: 1 year: 2015 end-page: 196 article-title: Computing sparse representation in a highly coherent dictionary based on difference of L1 and L2 publication-title: J. Sci. Comput. – volume: 33 start-page: 250 issue: 1 year: 2011 end-page: 278 article-title: Alternating direction algorithms for L1‐problemsincompressivesensing publication-title: SIAM J. Sci. Comput. |
| SSID | ssj0056512 |
| Score | 2.1938162 |
| Snippet | In sparse signal recovery, to overcome the ℓ1 ‐norm sparse regularisation's disadvantages tendency of uniformly penalise the signal amplitude and underestimate... |
| SourceID | wiley |
| SourceType | Publisher |
| StartPage | 1091 |
| SubjectTerms | approximation theory cluster point concave programming difference‐of‐convex algorithm framework high‐amplitude components minimax techniques nonconvex minimax‐concave penalty nonconvex nonsmooth problem nonconvex problem reconstruction quality signal amplitude signal reconstruction sparse signal recovery ℓ1 ‐norm loss function ℓ1 ‐norm sparse regularisation |
| Title | Sparse signal recovery via minimax‐concave penalty and ℓ1 ‐norm loss function |
| URI | https://onlinelibrary.wiley.com/doi/abs/10.1049%2Fiet-spr.2018.5130 |
| Volume | 12 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVWIB databaseName: Wiley Online Library Free Content customDbUrl: eissn: 1751-9683 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0056512 issn: 1751-9683 databaseCode: WIN dateStart: 20130101 isFulltext: true titleUrlDefault: https://onlinelibrary.wiley.com providerName: Wiley-Blackwell – providerCode: PRVWIB databaseName: Wiley Online Library Open Access customDbUrl: eissn: 1751-9683 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0056512 issn: 1751-9683 databaseCode: 24P dateStart: 20130101 isFulltext: true titleUrlDefault: https://authorservices.wiley.com/open-science/open-access/browse-journals.html providerName: Wiley-Blackwell |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1NS8MwGA5zetCD3-I3OXiNrkmbj6MOhwMZgynuVtI0gYHWstXhbh78AYL_cL_EN-02FG_ipYc0DeHNm_ejb_I8CJ0pJRIuNCMcHAAJXcSIblBNAiGtdFymKjIl2YTodGS_r7o11JzfhanwIRY_3PzOKO213-A6qVhIIKiFRRzYgoxyD-kZyPMITPESWg4CJj1_Aw27c3MMAUtV8hSeT55LtihtqotfQ_yMUEsX09r4l8ltovVZhIkvK5XYQjWbbaO1b7iDO6jXyyGftdgf3oCuPikGjZ7g8UBjjzXypF-nbx-QKRs9tji30KmYYJ2lePr-GWB4l0Goix_BwWLvGP3i7qL71vVd84bM2BXIyMO6EaU1NbyhpZGcC2FtlBhqaeiMiKhynFsFRrSkKDOpTUMjaMid0qlJoY022B6qZ8-Z3UfYMhlRoxqpS9JQSZMwy6hxmlFHmTL8ALFSWHFeIWjEZeU7VDGIKQYxxV5MsRdT3Gvf0quWv7gSHP7pqyO06turEyfHqF4MX-wJWjHjYjAanpaqAc-HducLjSDBXw |
| linkProvider | Wiley-Blackwell |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV3JTsMwELVKQQIO7IgdH7gGGjvxcmSrWlEipBbRW-Q4jlQJQtSGit448AFI_GG_BE_SViBuiGs8saKX8Swe-w1CJ1LyiHFFHWYdgOMlPnVUjSjH5cKIhIlY-rpoNsGDQHS78q6CrqZ3YUp-iNmGG6yMwl7DAocN6TLh9IAks2dyZ5ABp6crTn1ri-fQvGcDDmjg8NAMpvbYRixlzZNDQ3km6Ky2Kc9-TfEzRC18TH31f75uDa1MYkx8XirFOqqYdAMtf2Me3ETtdmYzWoPh-IYVhbTY6vQID3sKA9vIk3odv33YXFmrocGZsUL5CKs0xuP3TxfbsdQGu_jRulgMrhF-7xa6r193LhvOpL-CMwBiN0cqRTSrKaEFY5wb40eaGOIlmvtEJowZac1o0aRMxyb2NCceS6SKdWyfkRrdRtX0OTU7CBsqfKJlLU6i2JNCR9RQohNFSUKo1GwX0QKtMCs5NMKi9u3J0MIUWphCgCkEmMJ2s0Uu6nB1xd3701vHaLHRuW2FrWZws4-WQKY8f3KAqnn_xRyiBT3Me4P-UaEnX_99xKg |
| linkToPdf | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV3LSgMxFA21iujCt_g2C7ejnSSTx1Ktg8VSClXobkjzgIKOQzsWu3PhBwj-Yb_EZKYtijtxO7kzhJOb-5ibnAvAmRCsR5nEAXUOICA2woGsIRmEjBtuKdciUkWzCdZq8W5XtCugPrsLU_JDzH-4-Z1R2Gu_wU2mbZlwEk-S2Td5MMw8p2fIzyNnixfAIolY6HUbkfbMHruIpax5Mt9QnnI8r22Ki1-f-BmiFj4mXv-f2W2AtWmMCS9LpdgEFZNugdVvzIPboNPJXEZroD--4UR9Wux0egxHfQk928iTfJ28fbhcWcmRgZlxQvkYylTDyftnCN1Y6oJd-OhcLPSu0S_vDniIb-6vb4Npf4Vg6IndAiElUrQmueKUMmZM1FPIIGIVi5CwlBrhzGjRpExpo4liiFArpFbaPUM1vAuq6XNq9gA0mEdIiZq2PU0EVz1sMFJWYmQRForuA1yglWQlh0ZS1L6JSBxMiYMp8TAlHqak02iiq9hfXQkP_vTWKVhu1-Ok2WjdHYIVL1IePzkC1XzwYo7Bkhrl_eHgpFCTL4M_w78 |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Sparse+signal+recovery+via+minimax%E2%80%90concave+penalty+and+%E2%84%931+%E2%80%90norm+loss+function&rft.jtitle=IET+signal+processing&rft.au=Sun%2C+Yuli&rft.au=Chen%2C+Hao&rft.au=Tao%2C+Jinxu&rft.date=2018-12-01&rft.pub=The+Institution+of+Engineering+and+Technology&rft.issn=1751-9683&rft.eissn=1751-9683&rft.volume=12&rft.issue=9&rft.spage=1091&rft.epage=1098&rft_id=info:doi/10.1049%2Fiet-spr.2018.5130&rft.externalDBID=10.1049%252Fiet-spr.2018.5130&rft.externalDocID=SIL2BF00681 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1751-9683&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1751-9683&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1751-9683&client=summon |