The Improvement of Dropout Strategy Based on Two Evolutionary Algorithms
Dropout strategy is a simple and common regularization method in the construction of deep network that it can control the status of units in the Dropout layers according to the constant probability values in the training processes to prevent the training from overfitting. However, the probability va...
Uložené v:
| Vydané v: | 2018 IEEE International Conference on Robotics and Biomimetics (ROBIO) s. 814 - 819 |
|---|---|
| Hlavní autori: | , , |
| Médium: | Konferenčný príspevok.. |
| Jazyk: | English |
| Vydavateľské údaje: |
IEEE
01.12.2018
|
| Predmet: | |
| On-line prístup: | Získať plný text |
| Tagy: |
Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
|
| Abstract | Dropout strategy is a simple and common regularization method in the construction of deep network that it can control the status of units in the Dropout layers according to the constant probability values in the training processes to prevent the training from overfitting. However, the probability values of the Dropout strategy are single and decided by users, which means that we need more training iterations to receive better results and avoid less fitting problem. In this paper, two evolutionary algorithms, genetic algorithm and differential evolution algorithm are used to optimize the set probability values of network units to improve dropout strategy and they are proved to be able to increase the accuracy of the original method to about 5%. |
|---|---|
| AbstractList | Dropout strategy is a simple and common regularization method in the construction of deep network that it can control the status of units in the Dropout layers according to the constant probability values in the training processes to prevent the training from overfitting. However, the probability values of the Dropout strategy are single and decided by users, which means that we need more training iterations to receive better results and avoid less fitting problem. In this paper, two evolutionary algorithms, genetic algorithm and differential evolution algorithm are used to optimize the set probability values of network units to improve dropout strategy and they are proved to be able to increase the accuracy of the original method to about 5%. |
| Author | Chen, Tianhao Jia, Wenchuan Sun, Yi |
| Author_xml | – sequence: 1 givenname: Tianhao surname: Chen fullname: Chen, Tianhao organization: Shanghai Key Laboratory of Intelligent Manufacturing and Robotics, School of Mechatronic Engineering and Automation, Shanghai University, Shanghai, P. R. China – sequence: 2 givenname: Wenchuan surname: Jia fullname: Jia, Wenchuan organization: Shanghai Key Laboratory of Intelligent Manufacturing and Robotics, School of Mechatronic Engineering and Automation, Shanghai University, Shanghai, P. R. China – sequence: 3 givenname: Yi surname: Sun fullname: Sun, Yi organization: School of Mechatronic Engineering and Automation, Shanghai University, Shanghai Key Laboratory of Intelligent Manufacturing and Robotics, Shanghai, P. R. China |
| BookMark | eNotj91OwjAYQGuiF4q8gN70BTb75_rtEhBlCckS3T3p2q-wZFuXUjC8vSZwde5Oznki92MYkZAXznLOWfn2XS-rOheMQw5F8S6FuiPzUgPXAjiTWsMj2TQHpNUwxXDGAcdEg6cfMUzhlOhPiibh_kKX5oiOhpE2v4Guz6E_pS6MJl7oot-H2KXDcHwmD970R5zfOCPN57pZbbJt_VWtFtusK1nKBNdSCKvRCVAcrWuFVWDBe6WkAKmMZG3BnLZCWdOiMQ69Q868AQdKyBl5vWo7RNxNsRv-M3a3PfkH1JtKqw |
| ContentType | Conference Proceeding |
| DBID | 6IE 6IL CBEJK RIE RIL |
| DOI | 10.1109/ROBIO.2018.8665324 |
| DatabaseName | IEEE Electronic Library (IEL) Conference Proceedings IEEE Xplore POP ALL IEEE Xplore All Conference Proceedings IEEE Xplore IEEE Proceedings Order Plans (POP All) 1998-Present |
| DatabaseTitleList | |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Statistics |
| EISBN | 9781728103778 1728103770 |
| EndPage | 819 |
| ExternalDocumentID | 8665324 |
| Genre | orig-research |
| GroupedDBID | 6IE 6IL CBEJK RIE RIL |
| ID | FETCH-LOGICAL-i90t-217322c7ed2841ecdb2c48c8ff4432834a30b60d7c24cabeaadefde10fa8d8423 |
| IEDL.DBID | RIE |
| IngestDate | Thu Jun 29 18:39:02 EDT 2023 |
| IsPeerReviewed | false |
| IsScholarly | false |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-i90t-217322c7ed2841ecdb2c48c8ff4432834a30b60d7c24cabeaadefde10fa8d8423 |
| PageCount | 6 |
| ParticipantIDs | ieee_primary_8665324 |
| PublicationCentury | 2000 |
| PublicationDate | 2018-Dec. |
| PublicationDateYYYYMMDD | 2018-12-01 |
| PublicationDate_xml | – month: 12 year: 2018 text: 2018-Dec. |
| PublicationDecade | 2010 |
| PublicationTitle | 2018 IEEE International Conference on Robotics and Biomimetics (ROBIO) |
| PublicationTitleAbbrev | ROBIO |
| PublicationYear | 2018 |
| Publisher | IEEE |
| Publisher_xml | – name: IEEE |
| Score | 1.6796509 |
| Snippet | Dropout strategy is a simple and common regularization method in the construction of deep network that it can control the status of units in the Dropout layers... |
| SourceID | ieee |
| SourceType | Publisher |
| StartPage | 814 |
| SubjectTerms | AI-Based Methods Big Data in Robotics and Automation Computer Vision for Other Robotic Applications Dropout strategy Encoding Evolutionary computation Genetic algorithms Mathematical model Sociology Statistics Training |
| Title | The Improvement of Dropout Strategy Based on Two Evolutionary Algorithms |
| URI | https://ieeexplore.ieee.org/document/8665324 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LTwIxEJ4A8cDJBxjf6cGjC1220O5RFIKJAWI4cCN9TJVEdw0uGP697e6KMfHirWmaNJl2Ml-nM98HcN1FHumOpgGKng4YlyqItekGMQ9RoO11Y52rljzy8VjM5_G0Aje7XhhEzIvPsOWH-V--SfXap8ranpvNAYAqVDnnRa_Wdx8MjdtPk_7DxBdriVa58JdiSh4whvv_2-oAmj-dd2S6iymHUMHkCOoeERaEyg0YuYMlRSogz-yR1JJ7L3WwzkhJNbslfRebDEkTMvtMyWBTXi-52pLb1-d0tcxe3j6aMBsOZnejoJRDCJYxzQL3dnDOpzkaF1FC1EZ1NBNaWMtY5EACkxFVPWq47jAtFUpp0BoMqZXCCIeajqGWpAmeAGFWoFSRc24bMkSmIsUEjazyzDBK0lNoeIss3gvCi0VpjLO_p8-h7o1e1HhcQC1brfES9vTGWWZ1lZ_SF0Cyl18 |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LTwIxEG4QTeTkA4xve_DoQne3y3aPohCICMRw4Eb6mCqJ7pp1wfDvbXdXjIkXb03TpMlMm_k6nfk-hK4DCH3pSeIAa0uHhlw4kVSBE4UuMNDtIJK5askwHI3YbBZNKuhm0wsDAHnxGTTtMP_LV4lc2lRZy3KzGQCwhbYDSj236Nb67oQhUetp3BmMbbkWa5ZLf2mm5CGjt_e_zfZR46f3Dk82UeUAVSA-RDWLCQtK5TrqG9fiIhmQ5_ZwovG9FTtYZrgkm13jjolOCicxnn4muLsqDxhP1_j29TlJF9nL20cDTXvd6V3fKQURnEVEMse8Hsz1kyEoE1NckEp4kjLJtKbUNzCBcp-INlGh9KjkAjhXoBW4RHOmmMFNR6gaJzEcI0w1Ay58c721SwGo8AVlxNfCcsMITk5Q3Vpk_l5QXsxLY5z-PX2FdvvTx-F8OBg9nKGadUBR8XGOqlm6hAu0I1fGSull7rEvTaqapg |
| 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%3Abook&rft.genre=proceeding&rft.title=2018+IEEE+International+Conference+on+Robotics+and+Biomimetics+%28ROBIO%29&rft.atitle=The+Improvement+of+Dropout+Strategy+Based+on+Two+Evolutionary+Algorithms&rft.au=Chen%2C+Tianhao&rft.au=Jia%2C+Wenchuan&rft.au=Sun%2C+Yi&rft.date=2018-12-01&rft.pub=IEEE&rft.spage=814&rft.epage=819&rft_id=info:doi/10.1109%2FROBIO.2018.8665324&rft.externalDocID=8665324 |