Multiobjective evolution of deep learning parameters for robot manipulator object recognition and grasping

Deep Learning (DL) is currently very popular because of its similarity to the hierarchical architecture of human brain with multiple levels of abstraction. DL has many parameters that influence the network performance. In this paper, we introduce a multiobjective evolutionary algorithm (MOEA) to opt...

Full description

Saved in:
Bibliographic Details
Published inAdvanced robotics Vol. 32; no. 20; pp. 1090 - 1101
Main Authors Hossain, Delowar, Capi, Genci
Format Journal Article
LanguageEnglish
Published Taylor & Francis 18.10.2018
Subjects
DBNN
Deep learning
multiobjective evolution
object recognition
robot grasping
Online AccessGet full text
ISSN0169-1864
1568-5535
DOI10.1080/01691864.2018.1529620

Cover

Abstract Deep Learning (DL) is currently very popular because of its similarity to the hierarchical architecture of human brain with multiple levels of abstraction. DL has many parameters that influence the network performance. In this paper, we introduce a multiobjective evolutionary algorithm (MOEA) to optimize the DBNN parameters subject to the error rate and the network training time as two conflicting objectives. To verify the effectiveness, the proposed method is applied to the robot object recognition and grasping task. We compare the performance of the optimized DBNN model with a) DBNN with arbitrarily selected parameters and b) Deep Belief Network-Deep Neural Network (DBN-DNN). The results show that optimized DL has a superior performance in terms of training time and recognition success rate. In addition, the optimized DBNN model is effective for real-time robotic implementations.
AbstractList Deep Learning (DL) is currently very popular because of its similarity to the hierarchical architecture of human brain with multiple levels of abstraction. DL has many parameters that influence the network performance. In this paper, we introduce a multiobjective evolutionary algorithm (MOEA) to optimize the DBNN parameters subject to the error rate and the network training time as two conflicting objectives. To verify the effectiveness, the proposed method is applied to the robot object recognition and grasping task. We compare the performance of the optimized DBNN model with a) DBNN with arbitrarily selected parameters and b) Deep Belief Network-Deep Neural Network (DBN-DNN). The results show that optimized DL has a superior performance in terms of training time and recognition success rate. In addition, the optimized DBNN model is effective for real-time robotic implementations.
Author Capi, Genci
Hossain, Delowar
Author_xml – sequence: 1
  givenname: Delowar
  surname: Hossain
  fullname: Hossain, Delowar
  organization: Assistive Robotics Laboratory, Department of Mechanical Engineering, Faculty of Science and Engineering, HOSEI University
– sequence: 2
  givenname: Genci
  surname: Capi
  fullname: Capi, Genci
  email: capi@hosei.ac.jp
  organization: Assistive Robotics Laboratory, Department of Mechanical Engineering, Faculty of Science and Engineering, HOSEI University
BookMark eNqFkM1KAzEUhYNUsK0-gpAXmJpMJvODG6X4BxU3uh7uJDclZZoMmbTSt3fG1o0LXV04nPNx-WZk4rxDQq45W3BWshvG84qXebZIGS8XXKZVnrIzMuUyLxMphZyQ6dhJxtIFmfX9hjFWZqKYks3rro3WNxtU0e6R4t63uyFw1BuqETvaIgRn3Zp2EGCLEUNPjQ80-MZHugVnu10LcUiOFBpQ-bWz3xBwmq4D9N0AuCTnBtoer053Tj4eH96Xz8nq7elleb9KlCjymAhujBKqLLkwoqowBSG1Vo00LGeFLhCblJnGZFKmXEudNrqAHAVCVrGCCzEnt0euCr7vA5pa2QjjOzGAbWvO6lFb_aOtHrXVJ23DWv5ad8FuIRz-3d0dd9YNdrbw6UOr6wiH1gcTwCnb1-JvxBeG44mb
CitedBy_id crossref_primary_10_1007_s12182_019_00417_w
crossref_primary_10_1016_j_asoc_2023_110412
crossref_primary_10_1016_j_neucom_2022_09_114
crossref_primary_10_1108_IR_09_2019_0180
crossref_primary_10_1109_TNNLS_2021_3100554
crossref_primary_10_1017_S0263574721000023
crossref_primary_10_1016_j_inffus_2020_10_014
crossref_primary_10_3390_s23135826
Cites_doi 10.1108/IR-05-2016-0140
10.1016/j.swevo.2011.03.001
10.1177/0278364914549607
10.1109/TEVC.2015.2424251
10.1109/TNNLS.2016.2582798
10.1109/TEVC.2007.892759
10.1023/B:VISI.0000029664.99615.94
10.1177/0278364917710318
10.1007/s11263-013-0620-5
10.14257/ijunesst.2015.8.11.31
10.1109/TNNLS.2013.2293418
10.1162/neco.2006.18.7.1527
10.1162/106365600568158
10.1109/4235.797969
10.1109/TGRS.2016.2543748
10.1016/0004-3702(77)90006-6
10.1109/ICCV.1999.790410
10.1109/TNNLS.2015.2469673
10.1109/TRO.2007.910773
ContentType Journal Article
Copyright 2018 Informa UK Limited, trading as Taylor & Francis Group and The Robotics Society of Japan 2018
Copyright_xml – notice: 2018 Informa UK Limited, trading as Taylor & Francis Group and The Robotics Society of Japan 2018
DBID AAYXX
CITATION
DOI 10.1080/01691864.2018.1529620
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
EISSN 1568-5535
EndPage 1101
ExternalDocumentID 10_1080_01691864_2018_1529620
1529620
Genre Article
GroupedDBID -~X
.QJ
0BK
0R~
23M
30N
4.4
5GY
AAENE
AAJMT
AALDU
AAMIU
AAPUL
AAQRR
ABCCY
ABDBF
ABFIM
ABJNI
ABLIJ
ABPAQ
ABPEM
ABTAI
ABXUL
ABXYU
ACGFS
ACTIO
ACUHS
ADCVX
ADGTB
ADMLS
ADYSH
AEISY
AENEX
AEOZL
AEPSL
AEVUW
AEYOC
AFRVT
AGDLA
AGMYJ
AHDZW
AIJEM
AIYEW
AKBVH
AKOOK
ALMA_UNASSIGNED_HOLDINGS
ALQZU
AMFWP
AMPGV
AQRUH
AVBZW
AWYRJ
BLEHA
CCCUG
CS3
DGEBU
DKSSO
EAP
EBS
EJD
EMK
EPL
EST
ESX
E~A
E~B
F5P
GTTXZ
H13
HF~
HZ~
H~P
I-F
IPNFZ
J.P
KYCEM
M4Z
NX~
O9-
P2P
RIG
RNANH
ROSJB
RTWRZ
S-T
SNACF
TBQAZ
TDBHL
TEN
TFL
TFT
TFW
TTHFI
TUROJ
TUS
UT5
ZGOLN
~S~
AAGDL
AAHIA
AAYXX
CITATION
ID FETCH-LOGICAL-c376t-31ffc3c8813f399e2a35ddcb5f0607d7eeb20fbf45521d5d2bd7a6e3ea4907133
ISSN 0169-1864
IngestDate Tue Jul 01 01:11:36 EDT 2025
Thu Apr 24 22:52:02 EDT 2025
Tue May 20 10:45:42 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue 20
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c376t-31ffc3c8813f399e2a35ddcb5f0607d7eeb20fbf45521d5d2bd7a6e3ea4907133
PageCount 12
ParticipantIDs informaworld_taylorfrancis_310_1080_01691864_2018_1529620
crossref_citationtrail_10_1080_01691864_2018_1529620
crossref_primary_10_1080_01691864_2018_1529620
ProviderPackageCode CITATION
AAYXX
PublicationCentury 2000
PublicationDate 2018-10-18
PublicationDateYYYYMMDD 2018-10-18
PublicationDate_xml – month: 10
  year: 2018
  text: 2018-10-18
  day: 18
PublicationDecade 2010
PublicationTitle Advanced robotics
PublicationYear 2018
Publisher Taylor & Francis
Publisher_xml – name: Taylor & Francis
References CIT0032
Domhan T (CIT0029) 2015
CIT0031
CIT0034
Agrawal P (CIT0006) 2014; 8695
Hossain D (CIT0030) 2017; 11
Bergstra J (CIT0027) 2012; 13
Liu H (CIT0014) 2017
CIT0036
CIT0013
CIT0035
CIT0038
CIT0015
CIT0037
CIT0018
CIT0017
Deb K (CIT0043) 2000; 1917
IFR (CIT0001) 2017
CIT0041
Deng H (CIT0026) 2017; 681
CIT0022
CIT0044
Coello CC (CIT0033) 2007
Bergstra JS (CIT0028) 2011
CIT0003
CIT0002
CIT0005
CIT0004
CIT0009
References_xml – volume: 8695
  start-page: 329
  year: 2014
  ident: CIT0006
  publication-title: Eur Conf Comput Vis
– ident: CIT0018
  doi: 10.1108/IR-05-2016-0140
– volume-title: World robotics industrial robots
  year: 2017
  ident: CIT0001
– start-page: 3460
  year: 2015
  ident: CIT0029
  publication-title: IJCAI'15 Proceedings of the 24th International Conference on Artificial Intelligence, Vol. 15; July; Buenos Aires, Argentina
– volume-title: Evolutionary algorithms for solving multi-objective problems
  year: 2007
  ident: CIT0033
– ident: CIT0035
  doi: 10.1016/j.swevo.2011.03.001
– ident: CIT0015
  doi: 10.1177/0278364914549607
– ident: CIT0036
  doi: 10.1109/TEVC.2015.2424251
– ident: CIT0041
  doi: 10.1109/TNNLS.2016.2582798
– ident: CIT0034
  doi: 10.1109/TEVC.2007.892759
– ident: CIT0004
  doi: 10.1023/B:VISI.0000029664.99615.94
– ident: CIT0017
  doi: 10.1177/0278364917710318
– ident: CIT0005
  doi: 10.1007/s11263-013-0620-5
– ident: CIT0022
  doi: 10.14257/ijunesst.2015.8.11.31
– volume: 13
  start-page: 281
  year: 2012
  ident: CIT0027
  publication-title: J Mach Learn Res
– ident: CIT0037
  doi: 10.1109/TNNLS.2013.2293418
– ident: CIT0009
  doi: 10.1162/neco.2006.18.7.1527
– ident: CIT0032
  doi: 10.1162/106365600568158
– volume: 681
  start-page: 417
  year: 2017
  ident: CIT0026
  publication-title: Bioins Comput Theor Appl
– ident: CIT0031
  doi: 10.1109/4235.797969
– ident: CIT0013
  doi: 10.1109/TGRS.2016.2543748
– volume: 1917
  start-page: 849
  year: 2000
  ident: CIT0043
  publication-title: Parallel Probl Solving Nat PPSN VI
– volume: 11
  start-page: 629
  issue: 3
  year: 2017
  ident: CIT0030
  publication-title: World Acad Sci, Eng Technol, Int J Mech, Aerosp, Ind, Mechatron Manuf Eng
– ident: CIT0002
  doi: 10.1016/0004-3702(77)90006-6
– start-page: 1
  issue: 99
  year: 2017
  ident: CIT0014
  publication-title: IEEE Trans Intell Transp Syst
– ident: CIT0003
  doi: 10.1109/ICCV.1999.790410
– start-page: 2546
  year: 2011
  ident: CIT0028
  publication-title: Advances in Neural Information Processing Systems; Granada, Spain
– ident: CIT0038
  doi: 10.1109/TNNLS.2015.2469673
– ident: CIT0044
  doi: 10.1109/TRO.2007.910773
SSID ssj0008437
Score 2.233308
Snippet Deep Learning (DL) is currently very popular because of its similarity to the hierarchical architecture of human brain with multiple levels of abstraction. DL...
SourceID crossref
informaworld
SourceType Enrichment Source
Index Database
Publisher
StartPage 1090
SubjectTerms DBNN
Deep learning
multiobjective evolution
object recognition
robot grasping
Title Multiobjective evolution of deep learning parameters for robot manipulator object recognition and grasping
URI https://www.tandfonline.com/doi/abs/10.1080/01691864.2018.1529620
Volume 32
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1Lb9swDBay9rIdiu6Fde0GHXYLHDiW7FjHYg8EA7ZTixU7zJAsqUjRxUHq7tBfX1IPR8GCPS9GoERSIH4hKYb8SMgbXmlRFMJmpeICLiiKZ4prkWEzeGkNY9LVcX_6XM3P-ceL8mI0-pZkLd32atLe7awr-RepwhjIFatk_0Kyw6IwAK9BvvAECcPzj2Tsqmc7deWV1tj8CLuhB6iNWcWeEJdjZPj-jpkvjn5hvO5U12Pm6sK174IRv8p4yCcKScqXa3mzitYtktXGtAG3SpItPweDKxdBjV1jPu7mH47Vwkfgl-0iDTRMHetr0I2uVO2nnh9pWLIS8FnPRz4xQZViTVfpyUiirt3EMvHKmyeaExNEEysMXsl0p4YPKZHI8QP7YW5ejT2cRBVW2ybPDu88IPsFXCNAce-fzt99_TLY6pp7VtX4_WONF7Kv79piy3vZ4rZNvJKzQ3IQrhP01GPjMRmZ5RPyKCGZfEqutlFCB5TQzlJECY0ooRuUUNiSOvnSBCXUr0ITlFBACY0oeUbOP7w_ezvPQoeNrAXD0oMBtrZlbV1PmQVP1RSSlVq3qrR5lc_0zBhV5FZZXoKXp0tdKD2TlWFGcuHCG8_J3rJbmheEMqaFKmUrGTY4kLI2eVuDTKXiUlZcHxEeT65pA_08dkG5bqaRpTYceIMH3oQDPyKTYdrK86_8boJIxdL0DrbWI7Zhv5z78j_mHpOHmx_NCdnr17fmFfirvXodQHcPi82TVg
linkProvider Library Specific Holdings
linkToHtml http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1JT8JAFJ4YPKgHdyOuc_BabDvTZY7GSFCBEyTcmllJUCmB4sFf77wupJioB65t3mTWt-W970PojoaK-T4zTiAoswGKoI6gijlABs-NJoTnfdy9ftgZ0pdRMKr1wkBZJcTQpgCKyHU1PG5IRlclcfeAIOLFIaREvBgYfFjo27B9O2BhBI-TuP2VNo5pgZtpRRyQqbp4fhtmzT6toZfW7E77AMlqxkW5yVtrmYmW_PoB5rjZkg7RfumW4ofiHh2hLT09Rns1sMITNMl7dVMxKVQk1p_lrcWpwUrrGS4ZKMYY8MQ_oM5mge3C8DwVaYYBaCMnC7NfilHwqnrJDmKnjsdzvoD-rVM0bD8NHjtOydTgSKugMqvIjZFExrFHjPV4tM9JoJQUgXFDN1KRtvG7a4ShgfUWVKB8oSIeaqI5ZXmYfIYa03SqzxEmRDERcMkJAOVzHmtXxi5zuaCch1Q1Ea3OJ5EljDmwabwnXoV2Wm5lAluZlFvZRK2V2KzA8fhPgNUPP8nyBIop2E4S8qfsxQayt2inM-h1k-5z__US7cIvMJVefIUa2Xypr60PlImb_JJ_AxRa-jc
linkToPdf http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV07T8MwELZQkRAMvBHl6YE1JYkd1x4RUPGsGEBii-zYrlSgqdqUgV-PL06qFgkYuiY6yz4793Duvg-hM8q0iGNhg0RR4RIURQNFtQiADF5aQ4gs-7gfu-zmhd69JnU14bgqq4Qc2nqgiNJWw8c91LauiDsHAJGIM7gRiTgQ-AgWu6x9mcFPPujiCLtTY8yph810IgHI1E08vw0z557mwEtn3E5nA6l6wr7a5K01KVQr-_qB5bjQijbRehWU4gt_irbQkhlso7UZqMId1C87dXPV9wYSm8_qzOLcYm3MEFf8Ez0MaOIfUGUzxm5deJSrvMAAs1FShbknfhQ8rV1yg7iZ495IjqF7axe9dK6fL2-CiqchyJx5KpwZtzYjGecRsS7eMbEkidaZSmzIwrZuG5e9h1ZZmrhYQSc6VrotmSFGUlEmyXuoMcgHZh9hQrRQicwkAZh8KbkJMx6KUCoqJaO6iWi9PWlWgZgDl8Z7GtVYp5UqU1BlWqmyiVpTsaFH8fhPQMzufVqU1yfWc52k5E_ZgwVkT9HK01Unfbjt3h-iVXgDfjLiR6hRjCbm2AVAhTopj_g3_5z42w
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=Multiobjective+evolution+of+deep+learning+parameters+for+robot+manipulator+object+recognition+and+grasping&rft.jtitle=Advanced+robotics&rft.au=Hossain%2C+Delowar&rft.au=Capi%2C+Genci&rft.date=2018-10-18&rft.pub=Taylor+%26+Francis&rft.issn=0169-1864&rft.eissn=1568-5535&rft.volume=32&rft.issue=20&rft.spage=1090&rft.epage=1101&rft_id=info:doi/10.1080%2F01691864.2018.1529620&rft.externalDocID=1529620
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0169-1864&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0169-1864&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0169-1864&client=summon