Beyond EM: A faster Bayesian linear regression algorithm without matrix inversions
•We propose a novel matrix inversion free algorithm for solving the Bayesian linear regression problem with Gaussian priors.•Compared with the conventional gradient based and EM algorithms, our proposed one runs significantly faster while achieving the similar precision performance.•An iterative met...
Saved in:
| Published in: | Neurocomputing (Amsterdam) Vol. 378; pp. 435 - 440 |
|---|---|
| Main Author: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
Elsevier B.V
22.02.2020
|
| Subjects: | |
| ISSN: | 0925-2312, 1872-8286 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | •We propose a novel matrix inversion free algorithm for solving the Bayesian linear regression problem with Gaussian priors.•Compared with the conventional gradient based and EM algorithms, our proposed one runs significantly faster while achieving the similar precision performance.•An iterative method from a perspective of latent matrices is proposed for optimal search without breaking the symmetric positive definite manifold constraint.
The Bayesian linear regression is a useful tool for many scientific communities. This paper presents a novel algorithm for solving the Bayesian linear regression problem with Gaussian priors, which shares the same spirit as the gradient based methods. In addition, the standard scheme for this task, the Expectation Maximization (EM) algorithm, involves matrix inversions but our proposed algorithm is free of. Numerical experiments demonstrate that the proposed algorithm performs as well as the gradient based and EM algorithms in term of precision, but runs significantly faster than the gradient based and EM algorithms. Due to its matrix-inversion-free nature, the algorithm of this paper is a viable alternative to the competing methods available in the literature. |
|---|---|
| AbstractList | •We propose a novel matrix inversion free algorithm for solving the Bayesian linear regression problem with Gaussian priors.•Compared with the conventional gradient based and EM algorithms, our proposed one runs significantly faster while achieving the similar precision performance.•An iterative method from a perspective of latent matrices is proposed for optimal search without breaking the symmetric positive definite manifold constraint.
The Bayesian linear regression is a useful tool for many scientific communities. This paper presents a novel algorithm for solving the Bayesian linear regression problem with Gaussian priors, which shares the same spirit as the gradient based methods. In addition, the standard scheme for this task, the Expectation Maximization (EM) algorithm, involves matrix inversions but our proposed algorithm is free of. Numerical experiments demonstrate that the proposed algorithm performs as well as the gradient based and EM algorithms in term of precision, but runs significantly faster than the gradient based and EM algorithms. Due to its matrix-inversion-free nature, the algorithm of this paper is a viable alternative to the competing methods available in the literature. |
| Author | Tang, Ying |
| Author_xml | – sequence: 1 givenname: Ying orcidid: 0000-0002-6063-4481 surname: Tang fullname: Tang, Ying email: mathtygo@gmail.com organization: College of Cyber Security, Chengdu University of Technology, Chengdu 610059, PR China |
| BookMark | eNqFkMtOwzAQRS1UJNrCH7DwDyT4kThJF0htVR5SERKCdeTak-IqsZHtFvr3JCorFrCZkUZzrmbOBI2ss4DQNSUpJVTc7FILe-W6lBFa9aOUCHqGxrQsWFKyUozQmFQsTxin7AJNQtgRQgvKqjF6WcDRWY1XTzM8x40METxeyCMEIy1ujQXpsYethxCMs1i2W-dNfO_wZ1_dPuJORm--sLEH8MNKuETnjWwDXP30KXq7W70uH5L18_3jcr5OFCciJlkjNSlJqYEQ0QhBeUF5k_OGaSoyWoqN0DlXG6ZBVrzJZA6Sa1koxoBqpvgUZadc5V0IHpr6w5tO-mNNST14qXf1yUs9eBmmvZcem_3ClIky9pdHL037H3x7gqF_7GDA10EZsAq08aBirZ35O-AbHRuEug |
| CitedBy_id | crossref_primary_10_1007_s10723_021_09561_3 crossref_primary_10_1016_j_ecoleng_2022_106742 crossref_primary_10_1134_S1054661820030074 crossref_primary_10_3390_genes14081579 |
| Cites_doi | 10.1109/MSP.2012.2205597 10.1109/JSAC.2013.130214 10.1137/S1052623496303470 10.1214/13-BA852 10.1109/TBME.2014.2351376 10.1109/TITS.2015.2453116 10.1016/j.automatica.2013.03.030 10.1109/TIP.2008.2007354 10.1109/TIT.1985.1057115 10.1109/TAC.2016.2582642 10.1111/j.0006-341X.2004.00184.x 10.1109/TAC.2015.2426291 10.1007/s11222-006-8769-1 |
| ContentType | Journal Article |
| Copyright | 2019 |
| Copyright_xml | – notice: 2019 |
| DBID | AAYXX CITATION |
| DOI | 10.1016/j.neucom.2019.10.061 |
| DatabaseName | CrossRef |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Computer Science |
| EISSN | 1872-8286 |
| EndPage | 440 |
| ExternalDocumentID | 10_1016_j_neucom_2019_10_061 S0925231219314997 |
| GroupedDBID | --- --K --M .DC .~1 0R~ 123 1B1 1~. 1~5 4.4 457 4G. 53G 5VS 7-5 71M 8P~ 9JM 9JN AABNK AACTN AADPK AAEDT AAEDW AAIAV AAIKJ AAKOC AALRI AAOAW AAQFI AAXLA AAXUO AAYFN ABBOA ABCQJ ABFNM ABJNI ABMAC ABYKQ ACDAQ ACGFS ACRLP ACZNC ADBBV ADEZE AEBSH AEKER AENEX AFKWA AFTJW AFXIZ AGHFR AGUBO AGWIK AGYEJ AHHHB AHZHX AIALX AIEXJ AIKHN AITUG AJOXV ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AOUOD AXJTR BKOJK BLXMC CS3 DU5 EBS EFJIC EFLBG EO8 EO9 EP2 EP3 F5P FDB FIRID FNPLU FYGXN G-Q GBLVA GBOLZ IHE J1W KOM LG9 M41 MO0 MOBAO N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. Q38 ROL RPZ SDF SDG SDP SES SPC SPCBC SSN SSV SSZ T5K ZMT ~G- 29N 9DU AAQXK AATTM AAXKI AAYWO AAYXX ABWVN ABXDB ACLOT ACNNM ACRPL ACVFH ADCNI ADJOM ADMUD ADNMO AEIPS AEUPX AFJKZ AFPUW AGQPQ AIGII AIIUN AKBMS AKRWK AKYEP ANKPU APXCP ASPBG AVWKF AZFZN CITATION EFKBS EJD FEDTE FGOYB HLZ HVGLF HZ~ R2- SBC SEW WUQ XPP ~HD |
| ID | FETCH-LOGICAL-c306t-4fad0808de006f6613713f53f2d164186b6d53cb2dea93f4a5ea3da7c22e1d2c3 |
| ISICitedReferencesCount | 5 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000506202200038&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 0925-2312 |
| IngestDate | Sat Nov 29 07:12:59 EST 2025 Tue Nov 18 20:40:24 EST 2025 Fri Feb 23 02:49:35 EST 2024 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Matrix-inversion-free Bayesian linear regression Gaussian prior |
| Language | English |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c306t-4fad0808de006f6613713f53f2d164186b6d53cb2dea93f4a5ea3da7c22e1d2c3 |
| ORCID | 0000-0002-6063-4481 |
| PageCount | 6 |
| ParticipantIDs | crossref_primary_10_1016_j_neucom_2019_10_061 crossref_citationtrail_10_1016_j_neucom_2019_10_061 elsevier_sciencedirect_doi_10_1016_j_neucom_2019_10_061 |
| PublicationCentury | 2000 |
| PublicationDate | 2020-02-22 |
| PublicationDateYYYYMMDD | 2020-02-22 |
| PublicationDate_xml | – month: 02 year: 2020 text: 2020-02-22 day: 22 |
| PublicationDecade | 2020 |
| PublicationTitle | Neurocomputing (Amsterdam) |
| PublicationYear | 2020 |
| Publisher | Elsevier B.V |
| Publisher_xml | – name: Elsevier B.V |
| References | Babacan, Molina, Katsaggelos (bib0002) 2009; 18 Boashash (bib0008) 2015 Golub, Van Loan (bib0020) 2012; 3 Pan, Yuan, Gonçalves, Stan (bib0013) 2016; 61 Chen, Ljung (bib0017) 2013; 49 Ter Braak (bib0015) 2006; 16 Nieto-Barajas, Contreras-Cristán (bib0009) 2014; 9 Lv, Duan, Kang, Li, Wang (bib0012) 2015; 16 Yin, Gesbert, Filippou, Liu (bib0004) 2013; 31 Kamper, Jansen, Goldwater (bib0006) 2015 Hinton, Deng, Yu, Dahl, Mohamed, Jaitly, Senior, Vanhoucke, Nguyen, Sainath (bib0005) 2012; 29 Levin, Weiss, Durand, Freeman (bib0003) 2009 Carli, Chen, Ljung (bib0014) 2017; 62 McLachlan, Krishnan (bib0019) 2007; 382 Le Gall (bib0021) 2014 Neelon, Dunson (bib0001) 2004; 60 Dürichen, Pimentel, Clifton, Schweikard, Clifton (bib0010) 2015; 62 Lagarias, Reeds, Wright, Wright (bib0016) 1998; 9 Dell’Acqua, Bellotti, Berta, De Gloria (bib0011) 2015; 16 Giles (bib0018) 2008 Tretter (bib0007) 1985; 31 Lagarias (10.1016/j.neucom.2019.10.061_bib0016) 1998; 9 Chen (10.1016/j.neucom.2019.10.061_bib0017) 2013; 49 Nieto-Barajas (10.1016/j.neucom.2019.10.061_bib0009) 2014; 9 Lv (10.1016/j.neucom.2019.10.061_bib0012) 2015; 16 Babacan (10.1016/j.neucom.2019.10.061_bib0002) 2009; 18 Tretter (10.1016/j.neucom.2019.10.061_bib0007) 1985; 31 Hinton (10.1016/j.neucom.2019.10.061_bib0005) 2012; 29 Carli (10.1016/j.neucom.2019.10.061_bib0014) 2017; 62 Neelon (10.1016/j.neucom.2019.10.061_bib0001) 2004; 60 Levin (10.1016/j.neucom.2019.10.061_bib0003) 2009 Le Gall (10.1016/j.neucom.2019.10.061_bib0021) 2014 Boashash (10.1016/j.neucom.2019.10.061_bib0008) 2015 Giles (10.1016/j.neucom.2019.10.061_bib0018) 2008 Dell’Acqua (10.1016/j.neucom.2019.10.061_bib0011) 2015; 16 Ter Braak (10.1016/j.neucom.2019.10.061_bib0015) 2006; 16 Dürichen (10.1016/j.neucom.2019.10.061_bib0010) 2015; 62 Pan (10.1016/j.neucom.2019.10.061_bib0013) 2016; 61 Golub (10.1016/j.neucom.2019.10.061_bib0020) 2012; 3 Yin (10.1016/j.neucom.2019.10.061_bib0004) 2013; 31 Kamper (10.1016/j.neucom.2019.10.061_bib0006) 2015 McLachlan (10.1016/j.neucom.2019.10.061_bib0019) 2007; 382 |
| References_xml | – volume: 31 start-page: 264 year: 2013 end-page: 273 ident: bib0004 article-title: A coordinated approach to channel estimation in large-scale multiple-antenna systems publication-title: IEEE J. Sel. Areas Commun. – volume: 9 start-page: 112 year: 1998 end-page: 147 ident: bib0016 article-title: Convergence properties of the Nelder–Mead simplex method in low dimensions publication-title: SIAM J. Optim. – volume: 61 start-page: 182 year: 2016 end-page: 187 ident: bib0013 article-title: A sparse Bayesian approach to the identification of nonlinear state-space systems publication-title: IEEE Trans. Autom. Contr. – start-page: 35 year: 2008 end-page: 44 ident: bib0018 article-title: Collected Matrix Derivative Results for Forward and Reverse Mode Algorithmic Differentiation publication-title: Proceedings of the Advances in Automatic Differentiation – start-page: 1964 year: 2009 end-page: 1971 ident: bib0003 article-title: Understanding and evaluating blind deconvolution algorithms publication-title: Proceedings of the CVPR 2009 – volume: 9 start-page: 147 year: 2014 end-page: 170 ident: bib0009 article-title: A Bayesian nonparametric approach for time series clustering publication-title: Bayesian Anal. – volume: 16 start-page: 239 year: 2006 end-page: 249 ident: bib0015 article-title: A markov chain monte carlo version of the genetic algorithm differential evolution: easy Bayesian computing for real parameter spaces publication-title: Stat. Comput. – start-page: 296 year: 2014 end-page: 303 ident: bib0021 article-title: Powers of tensors and fast matrix multiplication publication-title: Proceedings of the Thirty-ninth International Symposium on Symbolic and Algebraic Computation – volume: 31 start-page: 832 year: 1985 end-page: 835 ident: bib0007 article-title: Estimating the frequency of a noisy sinusoid by linear regression (corresp.) publication-title: IEEE Trans. Inf. Theory – volume: 16 start-page: 865 year: 2015 end-page: 873 ident: bib0012 article-title: Traffic flow prediction with big data: a deep learning approach publication-title: IEEE Trans. Intell. Transp. Syst. – volume: 29 start-page: 82 year: 2012 end-page: 97 ident: bib0005 article-title: Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups publication-title: IEEE Signal Process Mag. – volume: 3 year: 2012 ident: bib0020 article-title: Matrix Computations – volume: 18 start-page: 12 year: 2009 end-page: 26 ident: bib0002 article-title: Variational Bayesian blind deconvolution using a total variation prior publication-title: IEEE Trans. Image Process. – volume: 16 start-page: 3393 year: 2015 end-page: 3402 ident: bib0011 article-title: Time-aware multivariate nearest neighbor regression methods for traffic flow prediction publication-title: IEEE Trans. Intell. Transp. Syst. – volume: 62 start-page: 1471 year: 2017 end-page: 1477 ident: bib0014 article-title: Maximum entropy kernels for system identification publication-title: IEEE Trans. Autom. Contr. – volume: 49 start-page: 2213 year: 2013 end-page: 2220 ident: bib0017 article-title: Implementation of algorithms for tuning parameters in regularized least squares problems in system identification publication-title: Automatica – volume: 60 start-page: 398 year: 2004 end-page: 406 ident: bib0001 article-title: Bayesian isotonic regression and trend analysis publication-title: Biometrics – year: 2015 ident: bib0006 article-title: Fully unsupervised small-vocabulary speech recognition using a segmental Bayesian model publication-title: Proceedings of the Sixteenth Annual Conference of the International Speech Communication Association – year: 2015 ident: bib0008 article-title: Time-frequency Signal Analysis and Processing: A Comprehensive Reference – volume: 62 start-page: 314 year: 2015 end-page: 322 ident: bib0010 article-title: Multitask gaussian processes for multivariate physiological time-series analysis publication-title: IEEE Trans. Biomed. Eng. – volume: 382 year: 2007 ident: bib0019 article-title: The EM Algorithm and Extensions – volume: 29 start-page: 82 issue: 6 year: 2012 ident: 10.1016/j.neucom.2019.10.061_bib0005 article-title: Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups publication-title: IEEE Signal Process Mag. doi: 10.1109/MSP.2012.2205597 – start-page: 296 year: 2014 ident: 10.1016/j.neucom.2019.10.061_bib0021 article-title: Powers of tensors and fast matrix multiplication – volume: 31 start-page: 264 issue: 2 year: 2013 ident: 10.1016/j.neucom.2019.10.061_bib0004 article-title: A coordinated approach to channel estimation in large-scale multiple-antenna systems publication-title: IEEE J. Sel. Areas Commun. doi: 10.1109/JSAC.2013.130214 – year: 2015 ident: 10.1016/j.neucom.2019.10.061_bib0006 article-title: Fully unsupervised small-vocabulary speech recognition using a segmental Bayesian model – volume: 16 start-page: 865 issue: 2 year: 2015 ident: 10.1016/j.neucom.2019.10.061_bib0012 article-title: Traffic flow prediction with big data: a deep learning approach publication-title: IEEE Trans. Intell. Transp. Syst. – volume: 9 start-page: 112 issue: 1 year: 1998 ident: 10.1016/j.neucom.2019.10.061_bib0016 article-title: Convergence properties of the Nelder–Mead simplex method in low dimensions publication-title: SIAM J. Optim. doi: 10.1137/S1052623496303470 – volume: 9 start-page: 147 issue: 1 year: 2014 ident: 10.1016/j.neucom.2019.10.061_bib0009 article-title: A Bayesian nonparametric approach for time series clustering publication-title: Bayesian Anal. doi: 10.1214/13-BA852 – volume: 382 year: 2007 ident: 10.1016/j.neucom.2019.10.061_bib0019 – volume: 3 year: 2012 ident: 10.1016/j.neucom.2019.10.061_bib0020 – start-page: 1964 year: 2009 ident: 10.1016/j.neucom.2019.10.061_bib0003 article-title: Understanding and evaluating blind deconvolution algorithms – volume: 62 start-page: 314 issue: 1 year: 2015 ident: 10.1016/j.neucom.2019.10.061_bib0010 article-title: Multitask gaussian processes for multivariate physiological time-series analysis publication-title: IEEE Trans. Biomed. Eng. doi: 10.1109/TBME.2014.2351376 – year: 2015 ident: 10.1016/j.neucom.2019.10.061_bib0008 – volume: 16 start-page: 3393 issue: 6 year: 2015 ident: 10.1016/j.neucom.2019.10.061_bib0011 article-title: Time-aware multivariate nearest neighbor regression methods for traffic flow prediction publication-title: IEEE Trans. Intell. Transp. Syst. doi: 10.1109/TITS.2015.2453116 – volume: 49 start-page: 2213 issue: 7 year: 2013 ident: 10.1016/j.neucom.2019.10.061_bib0017 article-title: Implementation of algorithms for tuning parameters in regularized least squares problems in system identification publication-title: Automatica doi: 10.1016/j.automatica.2013.03.030 – volume: 18 start-page: 12 issue: 1 year: 2009 ident: 10.1016/j.neucom.2019.10.061_bib0002 article-title: Variational Bayesian blind deconvolution using a total variation prior publication-title: IEEE Trans. Image Process. doi: 10.1109/TIP.2008.2007354 – volume: 31 start-page: 832 issue: 6 year: 1985 ident: 10.1016/j.neucom.2019.10.061_bib0007 article-title: Estimating the frequency of a noisy sinusoid by linear regression (corresp.) publication-title: IEEE Trans. Inf. Theory doi: 10.1109/TIT.1985.1057115 – volume: 62 start-page: 1471 issue: 3 year: 2017 ident: 10.1016/j.neucom.2019.10.061_bib0014 article-title: Maximum entropy kernels for system identification publication-title: IEEE Trans. Autom. Contr. doi: 10.1109/TAC.2016.2582642 – volume: 60 start-page: 398 issue: 2 year: 2004 ident: 10.1016/j.neucom.2019.10.061_bib0001 article-title: Bayesian isotonic regression and trend analysis publication-title: Biometrics doi: 10.1111/j.0006-341X.2004.00184.x – start-page: 35 year: 2008 ident: 10.1016/j.neucom.2019.10.061_bib0018 article-title: Collected Matrix Derivative Results for Forward and Reverse Mode Algorithmic Differentiation – volume: 61 start-page: 182 issue: 1 year: 2016 ident: 10.1016/j.neucom.2019.10.061_bib0013 article-title: A sparse Bayesian approach to the identification of nonlinear state-space systems publication-title: IEEE Trans. Autom. Contr. doi: 10.1109/TAC.2015.2426291 – volume: 16 start-page: 239 issue: 3 year: 2006 ident: 10.1016/j.neucom.2019.10.061_bib0015 article-title: A markov chain monte carlo version of the genetic algorithm differential evolution: easy Bayesian computing for real parameter spaces publication-title: Stat. Comput. doi: 10.1007/s11222-006-8769-1 |
| SSID | ssj0017129 |
| Score | 2.3047366 |
| Snippet | •We propose a novel matrix inversion free algorithm for solving the Bayesian linear regression problem with Gaussian priors.•Compared with the conventional... |
| SourceID | crossref elsevier |
| SourceType | Enrichment Source Index Database Publisher |
| StartPage | 435 |
| SubjectTerms | Bayesian linear regression Gaussian prior Matrix-inversion-free |
| Title | Beyond EM: A faster Bayesian linear regression algorithm without matrix inversions |
| URI | https://dx.doi.org/10.1016/j.neucom.2019.10.061 |
| Volume | 378 |
| WOSCitedRecordID | wos000506202200038&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVESC databaseName: Elsevier SD Freedom Collection Journals 2021 customDbUrl: eissn: 1872-8286 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0017129 issn: 0925-2312 databaseCode: AIEXJ dateStart: 19950101 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3Na9swFBdbu8Mu7boP-rEOHXYLKrFkR9ZuoWR0g5axpZCdjCJLXUrjFict2X_fJz3ZCcvotsIuJggpFvpJT7_3_D4IeW_gEnSyq5hTxrIUOCvT0JGNQcE1yjhpMxeKTcizs3w0Ul9itc1ZKCcgqypfLNTNf4Ua2gBsHzr7D3C3fwoN8BtAhyfADs-_Aj7GpAxOMejc6VnIg6h_2hAv6Wmlrju1vUAP2Kqjry6u68n8xzQYZb2f8tTn7V90JtUdWtNmqxQ2pPMwoRhENDP0p_4Vpd9arVlhGO3Q35urMVoWQI30kdpLPXQ95AXthjxjQAqxn0WpmUse4tFXxarA0jxRMKaYlCTesSmmaFoT32hJuDyq7K335QFyoo687x3ma_8lMfY3PxU_E-CgoOcp-ZRscpkpkG2b_U-D0ef2a5JMOOZcjFNvQiiDn9_6u35PUVZox_AF2Yr6Au0jzjvkia1eku2mFgeNovkV-Yqw08HpB9qnCDptQKcIOl2CTlvQaQSdIuh0Cfprcv5xMDw-YbFeBjOg-M1Z6nQJCkBeWhClDoiXkIlwmXC8BKU4yXvjXpkJM-al1Uq4VGdWi1JLw7lNSm7EG7JRXVd2l9AesEg4rl0ltE219UntUgOaMpxfYUzS3SOiWaLCxGTyvqbJVdF4DV4WuLCFX1jfCgu7R1g76gaTqfyhv2xWv4iEEIleARvmwZH7jx55QJ4vz8JbsjGvb-0heWbu5pNZ_S7urHtoWYc2 |
| linkProvider | Elsevier |
| 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=Beyond+EM%3A+A+faster+Bayesian+linear+regression+algorithm+without+matrix+inversions&rft.jtitle=Neurocomputing+%28Amsterdam%29&rft.au=Tang%2C+Ying&rft.date=2020-02-22&rft.pub=Elsevier+B.V&rft.issn=0925-2312&rft.eissn=1872-8286&rft.volume=378&rft.spage=435&rft.epage=440&rft_id=info:doi/10.1016%2Fj.neucom.2019.10.061&rft.externalDocID=S0925231219314997 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0925-2312&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0925-2312&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0925-2312&client=summon |