Evolving simple and accurate symbolic regression models via asynchronous parallel computing
In machine learning, reducing the complexity of a model can help to improve its computational efficiency and avoid overfitting. In genetic programming (GP), the model complexity reduction is often achieved by reducing the size of evolved expressions. However, previous studies have demonstrated that...
Uloženo v:
| Vydáno v: | Applied soft computing Ročník 104; s. 107198 |
|---|---|
| Hlavní autoři: | , , , , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
Elsevier B.V
01.06.2021
|
| Témata: | |
| ISSN: | 1568-4946, 1872-9681 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Abstract | In machine learning, reducing the complexity of a model can help to improve its computational efficiency and avoid overfitting. In genetic programming (GP), the model complexity reduction is often achieved by reducing the size of evolved expressions. However, previous studies have demonstrated that the expression size reduction does not necessarily prevent model overfitting. Therefore, this paper uses the evaluation time – the computational time required to evaluate a GP model on data – as the estimate of model complexity. The evaluation time depends not only on the size of evolved expressions but also their composition, thus acting as a more nuanced measure of model complexity than the expression size alone. To discourage complexity, this study employs a novel method called asynchronous parallel GP (APGP) that introduces a race condition in the evolutionary process of GP; the race offers an evolutionary advantage to the simple solutions when their accuracy is competitive. To evaluate the proposed method, it is compared to the standard GP (GP) and GP with bloat control (GP+BC) methods on six challenging symbolic regression problems. APGP produced models that are significantly more accurate (on 6/6 problems) than those produced by both GP and GP+BC. In terms of complexity control, APGP prevailed over GP but not over GP+BC; however, GP+BC produced simpler solutions at the cost of test-set accuracy. Moreover, APGP took a significantly lower number of evaluations than both GP and GP+BC to meet a target training fitness in all tests. Our analysis of the proposed APGP also involved: (1) an ablation study that separated the proposed measure of complexity from the race condition in APGP and (2) the study of an initialisation scheme that encourages functional diversity in the initial population that improved the results for all the GP methods. These results question the overall benefits of bloat control and endorse the employment of both the evaluation time as an estimate of model complexity and the proposed APGP method for controlling it.
•Managing the complexity of genetic programming models is an ongoing challenge.•The time it takes to evaluate a model with data can indicate its complexity.•Evaluation time can reflect a model’s size, computational and functional complexity.•Putting models in a race to complete evaluations is an untried idea to contain complexity.•The proposed Genetic Programming method is useful for symbolic regression and beyond. |
|---|---|
| AbstractList | In machine learning, reducing the complexity of a model can help to improve its computational efficiency and avoid overfitting. In genetic programming (GP), the model complexity reduction is often achieved by reducing the size of evolved expressions. However, previous studies have demonstrated that the expression size reduction does not necessarily prevent model overfitting. Therefore, this paper uses the evaluation time – the computational time required to evaluate a GP model on data – as the estimate of model complexity. The evaluation time depends not only on the size of evolved expressions but also their composition, thus acting as a more nuanced measure of model complexity than the expression size alone. To discourage complexity, this study employs a novel method called asynchronous parallel GP (APGP) that introduces a race condition in the evolutionary process of GP; the race offers an evolutionary advantage to the simple solutions when their accuracy is competitive. To evaluate the proposed method, it is compared to the standard GP (GP) and GP with bloat control (GP+BC) methods on six challenging symbolic regression problems. APGP produced models that are significantly more accurate (on 6/6 problems) than those produced by both GP and GP+BC. In terms of complexity control, APGP prevailed over GP but not over GP+BC; however, GP+BC produced simpler solutions at the cost of test-set accuracy. Moreover, APGP took a significantly lower number of evaluations than both GP and GP+BC to meet a target training fitness in all tests. Our analysis of the proposed APGP also involved: (1) an ablation study that separated the proposed measure of complexity from the race condition in APGP and (2) the study of an initialisation scheme that encourages functional diversity in the initial population that improved the results for all the GP methods. These results question the overall benefits of bloat control and endorse the employment of both the evaluation time as an estimate of model complexity and the proposed APGP method for controlling it.
•Managing the complexity of genetic programming models is an ongoing challenge.•The time it takes to evaluate a model with data can indicate its complexity.•Evaluation time can reflect a model’s size, computational and functional complexity.•Putting models in a race to complete evaluations is an untried idea to contain complexity.•The proposed Genetic Programming method is useful for symbolic regression and beyond. |
| ArticleNumber | 107198 |
| Author | Sambo, Aliyu Sani Indramohan, Vivek Padmanaabhan Shah, Hanifa Azad, R. Muhammad Atif Kovalchuk, Yevgeniya |
| Author_xml | – sequence: 1 givenname: Aliyu Sani surname: Sambo fullname: Sambo, Aliyu Sani email: aliyu.sambo@mail.bcu.ac.uk organization: School of Computing and Digital Technology, Birmingham City University, UK – sequence: 2 givenname: R. Muhammad Atif orcidid: 0000-0002-4013-5415 surname: Azad fullname: Azad, R. Muhammad Atif organization: School of Computing and Digital Technology, Birmingham City University, UK – sequence: 3 givenname: Yevgeniya orcidid: 0000-0003-4306-4680 surname: Kovalchuk fullname: Kovalchuk, Yevgeniya organization: School of Computing and Digital Technology, Birmingham City University, UK – sequence: 4 givenname: Vivek Padmanaabhan surname: Indramohan fullname: Indramohan, Vivek Padmanaabhan organization: School of Health, Education and Life Sciences, Birmingham City University, UK – sequence: 5 givenname: Hanifa orcidid: 0000-0001-6289-9160 surname: Shah fullname: Shah, Hanifa organization: Faculty of Computing, Engineering and the Built Environment, Birmingham City University, UK |
| BookMark | eNp9kM1KAzEUhYNUsK2-gKu8wNRJ5i8DbqTUHyi40ZWLcCe5U1MySUmmA317U-rKhat7OPAduN-CzJx3SMg9y1csZ_XDfgXRqxXPOUtFw1pxReZMNDxra8FmKVe1yMq2rG_IIsZ9nqCWizn52kzeTsbtaDTDwSIFpykodQwwIo2nofPWKBpwFzBG4x0dvEYb6WSAQjw59R2888dIDxDAWrRU-eFwHNPkLbnuwUa8-71L8vm8-Vi_Ztv3l7f10zZTRVmOmWKa90WtdN7rTiBC26a-KIu2QdSqR2x4BaxrWZ03ijcgRA9dAVVX6Krqq2JJ-GVXBR9jwF4eghkgnCTL5VmP3MuzHnnWIy96EiT-QMqMMKYPxwDG_o8-XtDkASeDQUZl0CnUJqAapfbmP_wH9B2GCw |
| CitedBy_id | crossref_primary_10_1108_K_01_2022_0049 crossref_primary_10_1145_3735634 crossref_primary_10_1016_j_asoc_2021_108106 crossref_primary_10_1016_j_asoc_2024_112239 crossref_primary_10_1007_s11831_023_09922_z crossref_primary_10_1016_j_asoc_2022_109854 |
| Cites_doi | 10.3390/e22040408 10.1162/evco.2006.14.3.309 10.1109/ACCESS.2020.2975753 10.1109/TEVC.2008.926486 10.1023/A:1014538503543 10.1162/EVCO_a_00111 10.1002/wics.179 10.1007/s10710-012-9159-4 10.1016/0005-1098(78)90005-5 10.1109/TEVC.2006.871252 10.1007/s10710-012-9177-2 10.1016/S0167-8191(97)00045-8 10.1017/S0890060408000127 10.1109/TPAMI.2002.1017616 10.1023/A:1010070616149 10.1007/s10710-011-9150-5 10.1145/3233231 10.1109/TEVC.2018.2881392 10.1070/RM1970v025n06ABEH001269 10.1109/TEVC.2014.2306994 10.1109/TEVC.2007.903549 10.1109/CEC48606.2020.9185771 |
| ContentType | Journal Article |
| Copyright | 2021 The Authors |
| Copyright_xml | – notice: 2021 The Authors |
| DBID | 6I. AAFTH AAYXX CITATION |
| DOI | 10.1016/j.asoc.2021.107198 |
| DatabaseName | ScienceDirect Open Access Titles Elsevier:ScienceDirect:Open Access CrossRef |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Computer Science |
| EISSN | 1872-9681 |
| ExternalDocumentID | 10_1016_j_asoc_2021_107198 S1568494621001216 |
| GroupedDBID | --K --M .DC .~1 0R~ 1B1 1~. 1~5 23M 4.4 457 4G. 53G 5GY 5VS 6I. 6J9 7-5 71M 8P~ AABNK AACTN AAEDT AAEDW AAFTH AAIAV AAIKJ AAKOC AALRI AAOAW AAQFI AAQXK AAXUO AAYFN ABBOA ABFNM ABFRF ABJNI ABMAC ABXDB ABYKQ ACDAQ ACGFO ACGFS ACNNM ACRLP ACZNC ADBBV ADEZE ADJOM ADMUD ADTZH AEBSH AECPX AEFWE AEKER AENEX AFKWA AFTJW AGHFR AGUBO AGYEJ AHJVU AHZHX AIALX AIEXJ AIKHN AITUG AJBFU AJOXV ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AOUOD ASPBG AVWKF AXJTR AZFZN BJAXD BKOJK BLXMC CS3 EBS EFJIC EFLBG EJD EO8 EO9 EP2 EP3 F5P FDB FEDTE FGOYB FIRID FNPLU FYGXN G-Q GBLVA GBOLZ HVGLF HZ~ IHE J1W JJJVA KOM M41 MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. Q38 R2- RIG ROL RPZ SDF SDG SES SEW SPC SPCBC SST SSV SSZ T5K UHS UNMZH ~G- 9DU AATTM AAXKI AAYWO AAYXX ABWVN ACLOT ACRPL ACVFH ADCNI ADNMO AEIPS AEUPX AFJKZ AFPUW AGQPQ AIGII AIIUN AKBMS AKRWK AKYEP ANKPU APXCP CITATION EFKBS ~HD |
| ID | FETCH-LOGICAL-c344t-c1d2f36cd0fdb8eea9934434397eedcfee725a1b91607c27a88fab3a5b3d55f53 |
| ISICitedReferencesCount | 9 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000641373300011&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1568-4946 |
| IngestDate | Sat Nov 29 07:01:53 EST 2025 Tue Nov 18 21:40:51 EST 2025 Fri Feb 23 02:40:57 EST 2024 |
| IsDoiOpenAccess | true |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Parallel computing Model complexity Genetic programming Evaluation time |
| Language | English |
| License | This is an open access article under the CC BY license. |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c344t-c1d2f36cd0fdb8eea9934434397eedcfee725a1b91607c27a88fab3a5b3d55f53 |
| ORCID | 0000-0003-4306-4680 0000-0002-4013-5415 0000-0001-6289-9160 |
| OpenAccessLink | https://dx.doi.org/10.1016/j.asoc.2021.107198 |
| ParticipantIDs | crossref_primary_10_1016_j_asoc_2021_107198 crossref_citationtrail_10_1016_j_asoc_2021_107198 elsevier_sciencedirect_doi_10_1016_j_asoc_2021_107198 |
| PublicationCentury | 2000 |
| PublicationDate | June 2021 2021-06-00 |
| PublicationDateYYYYMMDD | 2021-06-01 |
| PublicationDate_xml | – month: 06 year: 2021 text: June 2021 |
| PublicationDecade | 2020 |
| PublicationTitle | Applied soft computing |
| PublicationYear | 2021 |
| Publisher | Elsevier B.V |
| Publisher_xml | – name: Elsevier B.V |
| References | Keijzer (b64) 2003 Vapnik (b39) 1998 Cantú-Paz (b54) 1998; 10 Schraudolph, Grefenstette (b32) 1992 W.B. Langdon, Genetic Improvement of Genetic Programming, in: 2020 IEEE Congress on Evolutionary Computation (CEC), 2020, pp. 1–8 Sambo, Azad, Kovalchuk, Indramohan, Shah (b66) 2020 Poli (b19) 2003; 2610 McPhee, Jarvis, Crane (b23) 2004 Soule, Foster, Dickinson (b8) 1996 Spector, Robinson (b16) 2002; 3 Ekart, Nemeth (b20) 2001; 2 Dua, Karra Taniskidou (b58) 2017 Rivlin (b34) 1974 Nannen (b30) 2010 Castelli, Manzoni, Silva, Vanneschi (b35) 2011; 6621 Zvonkin, Levin (b28) 1970; 25 Vladislavleva, Smits, den Hertog (b33) 2009; 13 Koza, Andre (b49) 1995 Syswerda (b56) 1991; 1 Hu, Payne, Banzhaf, Moore (b17) 2012; 13 Hoai, McKay, Essam (b12) 2006; 10 de Vega, Olague, Lanza, Banzhaf, Goodman, Menendez-Clavijo, Martinez (b46) 2020; 8 Kolmogorov (b25) 1965; 1 Chennupati, Azad, Ryan (b15) 2015 Raymond, Chen, Xue, Zhang (b42) 2019 Silva, Dignum, Vanneschi (b63) 2012; 13 Paris, Robilliard, Fonlupt (b1) 2003; 2936 Ni, Rockett (b44) 2014; 19 Dignum, Poli (b22) 2008 Vanneschi, Castelli, Silva (b9) 2010 Luke, Panait (b60) 2006; 14 (b13) 2018 Chen, Zhang, Xue (b40) 2019; 23 Hatwell, Gaber, Azad (b4) 2020 Koza (b2) 1992 . Gustafson, Burke, Krasnogor (b59) 2005; 1 Couture (b6) 2007 Kim, Kim, Yoo (b51) 2017; 10452 Luke, Panait (b61) 2002 Azad (b14) 2003 Dignum, Poli (b62) 2008; 4971 Azad, Ryan (b10) 2014; 22 Azad, Ryan (b43) 2011 Oussaidène, Chopard, Pictet, Tomassini (b52) 1997; 23 Scott, De Jong (b53) 2016 Kumar, Goyal, Varma (b5) 2017; 70 Vapnik (b38) 2013 Scott, De Jong (b50) 2015 Chen, Zhang, Cue (b41) 2016 Silva, Dignum, Vanneschi (b24) 2012; 13 Power, Ryan, Azad (b55) 2005; 2 Luke, Panait (b21) 2006; 14 Walker, Miller (b18) 2008; 12 Kulkarni, Harman (b36) 2011; 3 Koza (b7) 1992 Vapnik (b37) 1998 Rissanen (b29) 1978; 14 C. Simpson, J. Jewett, S. Turnbull, V. Stinner, PEP 418: Add monotonic time, performance counter, and process time functions, Website Lipton (b3) 2018; 61 White, McDermott, Castelli, Manzoni, Goldman, Kronberger, Jaskowski, O’Reilly, Luke (b57) 2013; 14 Vitányi (b27) 2020; 22 Iba, de Garis, Sato (b31) 1994 Koza (b11) 2008; 22 Kanungo, Mount, Netanyahu, Piatko, Silverman, Wu (b65) 2002; 24 Sambo, Azad, Kovalchuk, Indramohan, Shah (b45) 2020 Cover, Thomas (b26) 2006 Cantú-Paz (10.1016/j.asoc.2021.107198_b54) 1998; 10 Chen (10.1016/j.asoc.2021.107198_b41) 2016 Gustafson (10.1016/j.asoc.2021.107198_b59) 2005; 1 Silva (10.1016/j.asoc.2021.107198_b63) 2012; 13 Couture (10.1016/j.asoc.2021.107198_b6) 2007 Iba (10.1016/j.asoc.2021.107198_b31) 1994 Luke (10.1016/j.asoc.2021.107198_b21) 2006; 14 Vapnik (10.1016/j.asoc.2021.107198_b38) 2013 (10.1016/j.asoc.2021.107198_b13) 2018 Dua (10.1016/j.asoc.2021.107198_b58) 2017 Kanungo (10.1016/j.asoc.2021.107198_b65) 2002; 24 Azad (10.1016/j.asoc.2021.107198_b43) 2011 Sambo (10.1016/j.asoc.2021.107198_b45) 2020 Spector (10.1016/j.asoc.2021.107198_b16) 2002; 3 Walker (10.1016/j.asoc.2021.107198_b18) 2008; 12 McPhee (10.1016/j.asoc.2021.107198_b23) 2004 Vapnik (10.1016/j.asoc.2021.107198_b37) 1998 Kumar (10.1016/j.asoc.2021.107198_b5) 2017; 70 Koza (10.1016/j.asoc.2021.107198_b11) 2008; 22 Oussaidène (10.1016/j.asoc.2021.107198_b52) 1997; 23 Kolmogorov (10.1016/j.asoc.2021.107198_b25) 1965; 1 Koza (10.1016/j.asoc.2021.107198_b2) 1992 Dignum (10.1016/j.asoc.2021.107198_b62) 2008; 4971 Lipton (10.1016/j.asoc.2021.107198_b3) 2018; 61 Kim (10.1016/j.asoc.2021.107198_b51) 2017; 10452 Luke (10.1016/j.asoc.2021.107198_b61) 2002 Luke (10.1016/j.asoc.2021.107198_b60) 2006; 14 Hu (10.1016/j.asoc.2021.107198_b17) 2012; 13 Azad (10.1016/j.asoc.2021.107198_b14) 2003 Vitányi (10.1016/j.asoc.2021.107198_b27) 2020; 22 Power (10.1016/j.asoc.2021.107198_b55) 2005; 2 Cover (10.1016/j.asoc.2021.107198_b26) 2006 Vladislavleva (10.1016/j.asoc.2021.107198_b33) 2009; 13 Vapnik (10.1016/j.asoc.2021.107198_b39) 1998 de Vega (10.1016/j.asoc.2021.107198_b46) 2020; 8 Keijzer (10.1016/j.asoc.2021.107198_b64) 2003 Scott (10.1016/j.asoc.2021.107198_b53) 2016 Poli (10.1016/j.asoc.2021.107198_b19) 2003; 2610 Hoai (10.1016/j.asoc.2021.107198_b12) 2006; 10 Ni (10.1016/j.asoc.2021.107198_b44) 2014; 19 Vanneschi (10.1016/j.asoc.2021.107198_b9) 2010 Rivlin (10.1016/j.asoc.2021.107198_b34) 1974 Koza (10.1016/j.asoc.2021.107198_b49) 1995 Zvonkin (10.1016/j.asoc.2021.107198_b28) 1970; 25 Soule (10.1016/j.asoc.2021.107198_b8) 1996 Dignum (10.1016/j.asoc.2021.107198_b22) 2008 Nannen (10.1016/j.asoc.2021.107198_b30) 2010 Chen (10.1016/j.asoc.2021.107198_b40) 2019; 23 Syswerda (10.1016/j.asoc.2021.107198_b56) 1991; 1 10.1016/j.asoc.2021.107198_b47 Kulkarni (10.1016/j.asoc.2021.107198_b36) 2011; 3 Scott (10.1016/j.asoc.2021.107198_b50) 2015 10.1016/j.asoc.2021.107198_b48 Koza (10.1016/j.asoc.2021.107198_b7) 1992 Silva (10.1016/j.asoc.2021.107198_b24) 2012; 13 Castelli (10.1016/j.asoc.2021.107198_b35) 2011; 6621 Raymond (10.1016/j.asoc.2021.107198_b42) 2019 Hatwell (10.1016/j.asoc.2021.107198_b4) 2020 Rissanen (10.1016/j.asoc.2021.107198_b29) 1978; 14 Schraudolph (10.1016/j.asoc.2021.107198_b32) 1992 Chennupati (10.1016/j.asoc.2021.107198_b15) 2015 Ekart (10.1016/j.asoc.2021.107198_b20) 2001; 2 White (10.1016/j.asoc.2021.107198_b57) 2013; 14 Sambo (10.1016/j.asoc.2021.107198_b66) 2020 Paris (10.1016/j.asoc.2021.107198_b1) 2003; 2936 Azad (10.1016/j.asoc.2021.107198_b10) 2014; 22 |
| References_xml | – volume: 1 start-page: 912 year: 2005 end-page: 919 ident: b59 article-title: On improving genetic programming for symbolic regression publication-title: Proceedings of the 2005 IEEE Congress on Evolutionary Computation – volume: 6621 start-page: 25 year: 2011 end-page: 36 ident: b35 article-title: A quantitative study of learning and generalization in genetic programming publication-title: Proceedings of the 14th European Conference on Genetic Programming, EuroGP 2011 – volume: 25 start-page: 83 year: 1970 ident: b28 article-title: The complexity of finite objects and the development of the concepts of information and randomness by means of the theory of algorithms publication-title: Russian Math. Surveys – volume: 14 start-page: 465 year: 1978 end-page: 471 ident: b29 article-title: Modeling by shortest data description publication-title: Automatica – volume: 2936 start-page: 267 year: 2003 end-page: 277 ident: b1 article-title: Exploring overfitting in genetic programming publication-title: Evolution Artificielle, 6th International Conference – volume: 23 start-page: 1183 year: 1997 end-page: 1198 ident: b52 article-title: Parallel genetic programming and its application to trading model induction publication-title: Parallel Comput. – start-page: 1 year: 2020 end-page: 42 ident: b4 article-title: CHIRPS: Explaining random forest classification publication-title: Artif. Intell. Rev. – year: 1992 ident: b2 article-title: Genetic programming: On the programming of computers by means of natural selection – start-page: 265 year: 1994 end-page: 284 ident: b31 article-title: Genetic programming using a minimum description length principle publication-title: Advances in Genetic Programming – volume: 2610 start-page: 204 year: 2003 end-page: 217 ident: b19 article-title: A simple but theoretically-motivated method to control bloat in genetic programming publication-title: Genetic Programming, Proceedings of EuroGP’2003 – reference: C. Simpson, J. Jewett, S. Turnbull, V. Stinner, PEP 418: Add monotonic time, performance counter, and process time functions, Website, – year: 2007 ident: b6 article-title: Complexity and Chaos-State-Of-The-Art; Formulations and Measures of Complexity – year: 2003 ident: b14 article-title: A Position Independent Representation for Evolutionary Automatic Programming Algorithms - The Chorus System – start-page: 1315 year: 2011 end-page: 1322 ident: b43 article-title: Variance based selection to improve test set performance in genetic programming publication-title: GECCO ’11: Proceedings of the 13th Annual Conference on Genetic and Evolutionary Computation – start-page: 709 year: 2016 end-page: 716 ident: b41 article-title: Improving generalisation of genetic programming for symbolic regression with structural risk minimisation publication-title: GECCO ’16: Proceedings of the 2016 Annual Conference on Genetic and Evolutionary Computation – start-page: 1209 year: 2015 end-page: 1212 ident: b50 article-title: Evaluation-time bias in asynchronous evolutionary algorithms publication-title: GECCO’15 Student Workshop – volume: 14 start-page: 309 year: 2006 end-page: 344 ident: b21 article-title: A comparison of bloat control methods for genetic programming publication-title: Evol. Comput. – volume: 2 start-page: 61 year: 2001 end-page: 73 ident: b20 article-title: Selection based on the Pareto nondomination criterion for controlling code growth in genetic programming publication-title: Genet. Programm. Evol. Mach. – volume: 3 start-page: 7 year: 2002 end-page: 40 ident: b16 article-title: Genetic programming and autoconstructive evolution with the push programming language publication-title: Genet. Program. Evol. Mach. – year: 1974 ident: b34 publication-title: The Chebyshev Polynomials – start-page: 1007 year: 2015 end-page: 1014 ident: b15 article-title: Performance optimization of multi-core grammatical evolution generated parallel recursive programs publication-title: GECCO ’15: Proceedings of the 2015 Annual Conference on Genetic and Evolutionary Computation – volume: 19 start-page: 157 year: 2014 end-page: 166 ident: b44 article-title: Tikhonov regularization as a complexity measure in multiobjective genetic programming publication-title: IEEE Trans. Evol. Comput. – start-page: 1991 year: 1992 ident: b32 article-title: A user’s guide to GAucsd 1.4 publication-title: Computer Science & Engineering Department – volume: 70 start-page: 1935 year: 2017 end-page: 1944 ident: b5 article-title: Resource-efficient machine learning in 2 KB RAM for the internet of things publication-title: Proceedings of the 34th International Conference on Machine Learning – volume: 1 start-page: 94 year: 1991 end-page: 101 ident: b56 article-title: A study of reproduction in generational and steady-state genetic algorithms publication-title: Foundations of Genetic Algorithms – start-page: 2657 year: 2019 end-page: 2664 ident: b42 article-title: Genetic programming with rademacher complexity for symbolic regression publication-title: 2019 IEEE Congress on Evolutionary Computation, CEC 2019 – start-page: 877 year: 2010 end-page: 884 ident: b9 article-title: Measuring bloat, overfitting and functional complexity in genetic programming publication-title: GECCO ’10: Proceedings of the 12th Annual Conference on Genetic and Evolutionary Computation – year: 2010 ident: b30 article-title: A short introduction to model selection, Kolmogorov complexity and minimum description length (MDL) – year: 2013 ident: b38 publication-title: The Nature of Statistical Learning Theory – volume: 3 start-page: 543 year: 2011 end-page: 556 ident: b36 article-title: Statistical learning theory: a tutorial publication-title: Wiley Interdiscip. Rev. Comput. Stat. – start-page: 521 year: 2020 end-page: 528 ident: b45 article-title: Leveraging asynchronous parallel computing to produce simple genetic programming computational models publication-title: Proceedings of the 35th Annual ACM Symposium on Applied Computing – year: 1995 ident: b49 article-title: Parallel Genetic Programming on a Network of Transputers – start-page: 411 year: 2002 end-page: 421 ident: b61 article-title: Fighting bloat with nonparametric parsimony pressure publication-title: Parallel Problem Solving from Nature - PPSN VII – volume: 22 start-page: 185 year: 2008 end-page: 193 ident: b11 article-title: Human-competitive machine invention by means of genetic programming publication-title: Artif. Intell. Eng. Des. Anal. Manuf. – volume: 1 start-page: 1 year: 1965 end-page: 7 ident: b25 article-title: Three approaches to the quantitative definition ofinformation’ publication-title: Probl. Inf. Transm. – volume: 2 start-page: 1831 year: 2005 end-page: 1838 ident: b55 article-title: Promoting diversity using migration strategies in distributed genetic algorithms publication-title: 2005 IEEE Congress on Evolutionary Computation – volume: 13 start-page: 305 year: 2012 end-page: 337 ident: b17 article-title: Evolutionary dynamics on multiple scales: a quantitative analysis of the interplay between genotype, phenotype, and fitness in linear genetic programming publication-title: Genet. Program. Evol. Mach. – start-page: 195 year: 2020 end-page: 210 ident: b66 article-title: Time control or size control? Reducing complexity and improving accuracy of genetic programming models publication-title: European Conference on Genetic Programming (Part of EvoStar) – volume: 22 start-page: 408 year: 2020 ident: b27 article-title: How incomputable is Kolmogorov complexity? publication-title: Entropy – year: 1998 ident: b39 article-title: Statistical learning theory. 1998, Vol. 3 – start-page: 593 year: 2004 end-page: 604 ident: b23 article-title: On the strength of size limits in linear genetic programming publication-title: Genetic and Evolutionary Computation – GECCO 2004 – year: 1992 ident: b7 article-title: Genetic Programming: On the Programming of Computers by Means of Natural Selection – volume: 13 start-page: 197 year: 2012 end-page: 238 ident: b63 article-title: Operator equalisation for bloat free genetic programming and a survey of bloat control methods publication-title: Genet. Program. Evol. Mach. – volume: 13 start-page: 197 year: 2012 end-page: 238 ident: b24 article-title: Operator equalisation for bloat free genetic programming and a survey of bloat control methods publication-title: Genetic Program. Evol. Mach. – volume: 13 start-page: 333 year: 2009 end-page: 349 ident: b33 article-title: Order of nonlinearity as a complexity measure for models generated by symbolic regression via Pareto genetic programming publication-title: IEEE Trans. Evol. Comput. – start-page: 845 year: 2016 end-page: 852 ident: b53 article-title: Evaluation-time bias in quasi-generational and steady-state asynchronous evolutionary algorithms publication-title: GECCO ’16: Proceedings of the 2016 on Genetic and Evolutionary Computation Conference – start-page: 70 year: 2003 end-page: 82 ident: b64 article-title: Improving symbolic regression with interval arithmetic and linear scaling publication-title: European Conference on Genetic Programming – volume: 14 start-page: 309 year: 2006 end-page: 344 ident: b60 article-title: A comparison of bloat control methods for genetic programming publication-title: Evol. Comput. – volume: 22 start-page: 287 year: 2014 end-page: 317 ident: b10 article-title: A simple approach to lifetime learning in genetic programming based symbolic regression publication-title: Evol. Comput. – reference: W.B. Langdon, Genetic Improvement of Genetic Programming, in: 2020 IEEE Congress on Evolutionary Computation (CEC), 2020, pp. 1–8, – volume: 12 start-page: 397 year: 2008 end-page: 417 ident: b18 article-title: The automatic acquisition, evolution and reuse of modules in cartesian genetic programming publication-title: IEEE Trans. Evol. Comput. – volume: 4971 start-page: 110 year: 2008 end-page: 121 ident: b62 article-title: Operator equalisation and bloat free GP publication-title: Proceedings of the 11th European Conference on Genetic Programming, EuroGP 2008 – volume: 10 start-page: 157 year: 2006 end-page: 166 ident: b12 article-title: Representation and structural difficulty in genetic programming publication-title: IEEE Trans. Evol. Comput. – volume: 10452 start-page: 137 year: 2017 end-page: 142 ident: b51 article-title: GPGPGPU: Evaluation of parallelisation of genetic programming using GPGPU publication-title: Proceedings of the 9th International Symposium on Search Based Software Engineering, SSBSE 2017 – volume: 24 start-page: 881 year: 2002 end-page: 892 ident: b65 article-title: An efficient k-means clustering algorithm: analysis and implementation publication-title: IEEE Trans. Pattern Anal. Mach. Intell. – volume: 8 start-page: 38692 year: 2020 end-page: 38713 ident: b46 article-title: Time and individual duration in genetic programming publication-title: IEEE Access – volume: 14 start-page: 3 year: 2013 end-page: 29 ident: b57 article-title: Better GP benchmarks: community survey results and proposals publication-title: Genet. Program. Evol. Mach. – start-page: 158 year: 2008 end-page: 169 ident: b22 article-title: Crossover, sampling, bloat and the harmful effects of size limits publication-title: European Conference on Genetic Programming – year: 1998 ident: b37 publication-title: Statistical learning theory – volume: 10 start-page: 141 year: 1998 end-page: 171 ident: b54 article-title: A survey of parallel genetic algorithms publication-title: Calc. Paralleles Res. Syst. Repar. – reference: . – year: 2018 ident: b13 article-title: Handbook of Grammatical Evolution – volume: 23 start-page: 703 year: 2019 end-page: 717 ident: b40 article-title: Structural risk minimisation-driven genetic programming for enhancing generalisation in symbolic regression publication-title: IEEE Trans. Evol. Comput. – year: 2017 ident: b58 article-title: UCI Machine Learning Repository – start-page: 215 year: 1996 end-page: 223 ident: b8 article-title: Code growth in genetic programming publication-title: Genetic Programming 1996: Proceedings of the First Annual Conference – start-page: 16 year: 2006 ident: b26 article-title: Joint entropy and conditional entropy publication-title: Elements of Information Theory – volume: 61 start-page: 36 year: 2018 end-page: 43 ident: b3 article-title: The mythos of model interpretability publication-title: Commun. ACM – year: 2018 ident: 10.1016/j.asoc.2021.107198_b13 – volume: 22 start-page: 408 issue: 4 year: 2020 ident: 10.1016/j.asoc.2021.107198_b27 article-title: How incomputable is Kolmogorov complexity? publication-title: Entropy doi: 10.3390/e22040408 – volume: 1 start-page: 1 issue: 1 year: 1965 ident: 10.1016/j.asoc.2021.107198_b25 article-title: Three approaches to the quantitative definition ofinformation’ publication-title: Probl. Inf. Transm. – start-page: 1 year: 2020 ident: 10.1016/j.asoc.2021.107198_b4 article-title: CHIRPS: Explaining random forest classification publication-title: Artif. Intell. Rev. – year: 2017 ident: 10.1016/j.asoc.2021.107198_b58 – volume: 14 start-page: 309 issue: 3 year: 2006 ident: 10.1016/j.asoc.2021.107198_b60 article-title: A comparison of bloat control methods for genetic programming publication-title: Evol. Comput. doi: 10.1162/evco.2006.14.3.309 – volume: 8 start-page: 38692 year: 2020 ident: 10.1016/j.asoc.2021.107198_b46 article-title: Time and individual duration in genetic programming publication-title: IEEE Access doi: 10.1109/ACCESS.2020.2975753 – volume: 13 start-page: 333 issue: 2 year: 2009 ident: 10.1016/j.asoc.2021.107198_b33 article-title: Order of nonlinearity as a complexity measure for models generated by symbolic regression via Pareto genetic programming publication-title: IEEE Trans. Evol. Comput. doi: 10.1109/TEVC.2008.926486 – start-page: 16 year: 2006 ident: 10.1016/j.asoc.2021.107198_b26 article-title: Joint entropy and conditional entropy – year: 2007 ident: 10.1016/j.asoc.2021.107198_b6 – volume: 2 start-page: 1831 year: 2005 ident: 10.1016/j.asoc.2021.107198_b55 article-title: Promoting diversity using migration strategies in distributed genetic algorithms – start-page: 2657 year: 2019 ident: 10.1016/j.asoc.2021.107198_b42 article-title: Genetic programming with rademacher complexity for symbolic regression – start-page: 158 year: 2008 ident: 10.1016/j.asoc.2021.107198_b22 article-title: Crossover, sampling, bloat and the harmful effects of size limits – volume: 3 start-page: 7 issue: 1 year: 2002 ident: 10.1016/j.asoc.2021.107198_b16 article-title: Genetic programming and autoconstructive evolution with the push programming language publication-title: Genet. Program. Evol. Mach. doi: 10.1023/A:1014538503543 – year: 1998 ident: 10.1016/j.asoc.2021.107198_b37 – start-page: 1991 year: 1992 ident: 10.1016/j.asoc.2021.107198_b32 article-title: A user’s guide to GAucsd 1.4 – start-page: 845 year: 2016 ident: 10.1016/j.asoc.2021.107198_b53 article-title: Evaluation-time bias in quasi-generational and steady-state asynchronous evolutionary algorithms – volume: 2936 start-page: 267 year: 2003 ident: 10.1016/j.asoc.2021.107198_b1 article-title: Exploring overfitting in genetic programming – volume: 14 start-page: 309 issue: 3 year: 2006 ident: 10.1016/j.asoc.2021.107198_b21 article-title: A comparison of bloat control methods for genetic programming publication-title: Evol. Comput. doi: 10.1162/evco.2006.14.3.309 – volume: 22 start-page: 287 issue: 2 year: 2014 ident: 10.1016/j.asoc.2021.107198_b10 article-title: A simple approach to lifetime learning in genetic programming based symbolic regression publication-title: Evol. Comput. doi: 10.1162/EVCO_a_00111 – volume: 3 start-page: 543 issue: 6 year: 2011 ident: 10.1016/j.asoc.2021.107198_b36 article-title: Statistical learning theory: a tutorial publication-title: Wiley Interdiscip. Rev. Comput. Stat. doi: 10.1002/wics.179 – year: 1998 ident: 10.1016/j.asoc.2021.107198_b39 – start-page: 877 year: 2010 ident: 10.1016/j.asoc.2021.107198_b9 article-title: Measuring bloat, overfitting and functional complexity in genetic programming – start-page: 265 year: 1994 ident: 10.1016/j.asoc.2021.107198_b31 article-title: Genetic programming using a minimum description length principle – start-page: 1315 year: 2011 ident: 10.1016/j.asoc.2021.107198_b43 article-title: Variance based selection to improve test set performance in genetic programming – start-page: 195 year: 2020 ident: 10.1016/j.asoc.2021.107198_b66 article-title: Time control or size control? Reducing complexity and improving accuracy of genetic programming models – volume: 70 start-page: 1935 year: 2017 ident: 10.1016/j.asoc.2021.107198_b5 article-title: Resource-efficient machine learning in 2 KB RAM for the internet of things – year: 1992 ident: 10.1016/j.asoc.2021.107198_b7 – start-page: 521 year: 2020 ident: 10.1016/j.asoc.2021.107198_b45 article-title: Leveraging asynchronous parallel computing to produce simple genetic programming computational models – start-page: 1007 year: 2015 ident: 10.1016/j.asoc.2021.107198_b15 article-title: Performance optimization of multi-core grammatical evolution generated parallel recursive programs – volume: 13 start-page: 305 issue: 3 year: 2012 ident: 10.1016/j.asoc.2021.107198_b17 article-title: Evolutionary dynamics on multiple scales: a quantitative analysis of the interplay between genotype, phenotype, and fitness in linear genetic programming publication-title: Genet. Program. Evol. Mach. doi: 10.1007/s10710-012-9159-4 – volume: 14 start-page: 465 issue: 5 year: 1978 ident: 10.1016/j.asoc.2021.107198_b29 article-title: Modeling by shortest data description publication-title: Automatica doi: 10.1016/0005-1098(78)90005-5 – volume: 10 start-page: 157 issue: 2 year: 2006 ident: 10.1016/j.asoc.2021.107198_b12 article-title: Representation and structural difficulty in genetic programming publication-title: IEEE Trans. Evol. Comput. doi: 10.1109/TEVC.2006.871252 – volume: 14 start-page: 3 issue: 1 year: 2013 ident: 10.1016/j.asoc.2021.107198_b57 article-title: Better GP benchmarks: community survey results and proposals publication-title: Genet. Program. Evol. Mach. doi: 10.1007/s10710-012-9177-2 – volume: 4971 start-page: 110 year: 2008 ident: 10.1016/j.asoc.2021.107198_b62 article-title: Operator equalisation and bloat free GP – ident: 10.1016/j.asoc.2021.107198_b47 – volume: 1 start-page: 94 year: 1991 ident: 10.1016/j.asoc.2021.107198_b56 article-title: A study of reproduction in generational and steady-state genetic algorithms – year: 2003 ident: 10.1016/j.asoc.2021.107198_b14 – volume: 23 start-page: 1183 issue: 8 year: 1997 ident: 10.1016/j.asoc.2021.107198_b52 article-title: Parallel genetic programming and its application to trading model induction publication-title: Parallel Comput. doi: 10.1016/S0167-8191(97)00045-8 – volume: 22 start-page: 185 issue: 3 year: 2008 ident: 10.1016/j.asoc.2021.107198_b11 article-title: Human-competitive machine invention by means of genetic programming publication-title: Artif. Intell. Eng. Des. Anal. Manuf. doi: 10.1017/S0890060408000127 – volume: 10 start-page: 141 issue: 2 year: 1998 ident: 10.1016/j.asoc.2021.107198_b54 article-title: A survey of parallel genetic algorithms publication-title: Calc. Paralleles Res. Syst. Repar. – year: 1995 ident: 10.1016/j.asoc.2021.107198_b49 – volume: 10452 start-page: 137 year: 2017 ident: 10.1016/j.asoc.2021.107198_b51 article-title: GPGPGPU: Evaluation of parallelisation of genetic programming using GPGPU – volume: 24 start-page: 881 issue: 7 year: 2002 ident: 10.1016/j.asoc.2021.107198_b65 article-title: An efficient k-means clustering algorithm: analysis and implementation publication-title: IEEE Trans. Pattern Anal. Mach. Intell. doi: 10.1109/TPAMI.2002.1017616 – volume: 2610 start-page: 204 year: 2003 ident: 10.1016/j.asoc.2021.107198_b19 article-title: A simple but theoretically-motivated method to control bloat in genetic programming – year: 2010 ident: 10.1016/j.asoc.2021.107198_b30 – year: 1974 ident: 10.1016/j.asoc.2021.107198_b34 – start-page: 70 year: 2003 ident: 10.1016/j.asoc.2021.107198_b64 article-title: Improving symbolic regression with interval arithmetic and linear scaling – start-page: 1209 year: 2015 ident: 10.1016/j.asoc.2021.107198_b50 article-title: Evaluation-time bias in asynchronous evolutionary algorithms – volume: 2 start-page: 61 issue: 1 year: 2001 ident: 10.1016/j.asoc.2021.107198_b20 article-title: Selection based on the Pareto nondomination criterion for controlling code growth in genetic programming publication-title: Genet. Programm. Evol. Mach. doi: 10.1023/A:1010070616149 – start-page: 709 year: 2016 ident: 10.1016/j.asoc.2021.107198_b41 article-title: Improving generalisation of genetic programming for symbolic regression with structural risk minimisation – volume: 1 start-page: 912 year: 2005 ident: 10.1016/j.asoc.2021.107198_b59 article-title: On improving genetic programming for symbolic regression – start-page: 215 year: 1996 ident: 10.1016/j.asoc.2021.107198_b8 article-title: Code growth in genetic programming – volume: 13 start-page: 197 issue: 2 year: 2012 ident: 10.1016/j.asoc.2021.107198_b63 article-title: Operator equalisation for bloat free genetic programming and a survey of bloat control methods publication-title: Genet. Program. Evol. Mach. doi: 10.1007/s10710-011-9150-5 – start-page: 593 year: 2004 ident: 10.1016/j.asoc.2021.107198_b23 article-title: On the strength of size limits in linear genetic programming – year: 1992 ident: 10.1016/j.asoc.2021.107198_b2 – volume: 61 start-page: 36 issue: 10 year: 2018 ident: 10.1016/j.asoc.2021.107198_b3 article-title: The mythos of model interpretability publication-title: Commun. ACM doi: 10.1145/3233231 – start-page: 411 year: 2002 ident: 10.1016/j.asoc.2021.107198_b61 article-title: Fighting bloat with nonparametric parsimony pressure – volume: 23 start-page: 703 issue: 4 year: 2019 ident: 10.1016/j.asoc.2021.107198_b40 article-title: Structural risk minimisation-driven genetic programming for enhancing generalisation in symbolic regression publication-title: IEEE Trans. Evol. Comput. doi: 10.1109/TEVC.2018.2881392 – volume: 25 start-page: 83 issue: 6 year: 1970 ident: 10.1016/j.asoc.2021.107198_b28 article-title: The complexity of finite objects and the development of the concepts of information and randomness by means of the theory of algorithms publication-title: Russian Math. Surveys doi: 10.1070/RM1970v025n06ABEH001269 – volume: 19 start-page: 157 issue: 2 year: 2014 ident: 10.1016/j.asoc.2021.107198_b44 article-title: Tikhonov regularization as a complexity measure in multiobjective genetic programming publication-title: IEEE Trans. Evol. Comput. doi: 10.1109/TEVC.2014.2306994 – volume: 12 start-page: 397 issue: 4 year: 2008 ident: 10.1016/j.asoc.2021.107198_b18 article-title: The automatic acquisition, evolution and reuse of modules in cartesian genetic programming publication-title: IEEE Trans. Evol. Comput. doi: 10.1109/TEVC.2007.903549 – volume: 6621 start-page: 25 year: 2011 ident: 10.1016/j.asoc.2021.107198_b35 article-title: A quantitative study of learning and generalization in genetic programming – volume: 13 start-page: 197 issue: 2 year: 2012 ident: 10.1016/j.asoc.2021.107198_b24 article-title: Operator equalisation for bloat free genetic programming and a survey of bloat control methods publication-title: Genetic Program. Evol. Mach. doi: 10.1007/s10710-011-9150-5 – ident: 10.1016/j.asoc.2021.107198_b48 doi: 10.1109/CEC48606.2020.9185771 – year: 2013 ident: 10.1016/j.asoc.2021.107198_b38 |
| SSID | ssj0016928 |
| Score | 2.366641 |
| Snippet | In machine learning, reducing the complexity of a model can help to improve its computational efficiency and avoid overfitting. In genetic programming (GP),... |
| SourceID | crossref elsevier |
| SourceType | Enrichment Source Index Database Publisher |
| StartPage | 107198 |
| SubjectTerms | Evaluation time Genetic programming Model complexity Parallel computing |
| Title | Evolving simple and accurate symbolic regression models via asynchronous parallel computing |
| URI | https://dx.doi.org/10.1016/j.asoc.2021.107198 |
| Volume | 104 |
| WOSCitedRecordID | wos000641373300011&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: 1872-9681 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0016928 issn: 1568-4946 databaseCode: AIEXJ dateStart: 20010601 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1Nj9MwELVKlwMXlk-xLCAfuEWpmg839rFCRSwSKwQLKuIQObZDs5umqyaptvwUfi3j2k5LgRV74BJVkeOkmdfn8XTmDUIvqUoioYj0iciVH9Oc-TSSkT_ioQylYlJwuWk2kZye0umUve_1frhamFWZVBW9umKX_9XUcA6MrUtnb2DublI4AZ_B6HAEs8Pxnww_AcIxUYJCK_8aNVYhWq0J4dXreaaVgL2l-mYyYCvTDKf2Vro8q15XQsvl6sRYrQpelqrcpJ23jVvjnGStdV9r4PG9ETpgw-E-poCmWLfeR14VHbS-G1R9GHjv2hmfz7n0xk2Rd9S_gLchZu2Gpr-oFXzRYt0tHieVXPL5Ymbitp-Bqy_AC5Y6CZdnMwt1G8UId7KtHPGOqB8zG450zDyMd7gVNqqB6Vj9G-2bCMT5gAOiB3r6wXbwrxrbe2tfl5Hokt3OUz1HqudIzRy30EGYEEb76GB8Mpm-7f6jGrFN597uyW1Jlske3H-SP7s9O67M2T101-5B8Nhg5z7qqeoBOnT9PbCl-4foq4MSNlDCACXsoIQdlPAWSthACQOU8C6UsIMS7oDyCH16PTl79ca3zTh8EcVx44tAhnk0EnKYy4wqxcGxjXVZMkvAzYIfukpCwoOMacVCESac0pxnESdZJAnJSfQY9atFpZ4gDE6hUgy2ClTksRoGmQxonOs6JarIUNIjFLh3lQqrVK8bppTp3610hLzumkuj03LtaOJMkFpP03iQKSDqmuue3ugux-jOFurPUL9Ztuo5ui1WTVEvX1g4_QSeX6Xp |
| 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=Evolving+simple+and+accurate+symbolic+regression+models+via+asynchronous+parallel+computing&rft.jtitle=Applied+soft+computing&rft.au=Sambo%2C+Aliyu+Sani&rft.au=Azad%2C+R.+Muhammad+Atif&rft.au=Kovalchuk%2C+Yevgeniya&rft.au=Indramohan%2C+Vivek+Padmanaabhan&rft.date=2021-06-01&rft.issn=1568-4946&rft.volume=104&rft.spage=107198&rft_id=info:doi/10.1016%2Fj.asoc.2021.107198&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_asoc_2021_107198 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1568-4946&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1568-4946&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1568-4946&client=summon |