Generative autoencoder to prevent overregularization of variational autoencoder
In machine learning, data scarcity is a common problem, and generative models have the potential to solve it. The variational autoencoder is a generative model that performs variational inference to estimate a low‐dimensional posterior distribution given high‐dimensional data. Specifically, it optim...
Uloženo v:
| Vydáno v: | ETRI journal Ročník 47; číslo 1; s. 80 - 89 |
|---|---|
| Hlavní autoři: | , , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
Electronics and Telecommunications Research Institute (ETRI)
01.02.2025
한국전자통신연구원 |
| Témata: | |
| ISSN: | 1225-6463, 2233-7326 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Abstract | In machine learning, data scarcity is a common problem, and generative models have the potential to solve it. The variational autoencoder is a generative model that performs variational inference to estimate a low‐dimensional posterior distribution given high‐dimensional data. Specifically, it optimizes the evidence lower bound from regularization and reconstruction terms, but the two terms are imbalanced in general. If the reconstruction error is not sufficiently small to belong to the population, the generative model performance cannot be guaranteed. We propose a generative autoencoder (GAE) that uses an autoencoder to first minimize the reconstruction error and then estimate the distribution using latent vectors mapped onto a lower dimension through the encoder. We compare the Fréchet inception distances scores of the proposed GAE and nine other variational autoencoders on the MNIST, Fashion MNIST, CIFAR10, and SVHN datasets. The proposed GAE consistently outperforms the other methods on the MNIST (44.30), Fashion MNIST (196.34), and SVHN (77.53) datasets. |
|---|---|
| AbstractList | In machine learning, data scarcity is a common problem, and generative models have the potential to solve it. The variational autoencoder is a generative model that performs variational inference to estimate a low‐dimensional posterior distribution given high‐dimensional data. Specifically, it optimizes the evidence lower bound from regularization and reconstruction terms, but the two terms are imbalanced in general. If the reconstruction error is not sufficiently small to belong to the population, the generative model performance cannot be guaranteed. We propose a generative autoencoder (GAE) that uses an autoencoder to first minimize the reconstruction error and then estimate the distribution using latent vectors mapped onto a lower dimension through the encoder. We compare the Fréchet inception distances scores of the proposed GAE and nine other variational autoencoders on the MNIST, Fashion MNIST, CIFAR10, and SVHN datasets. The proposed GAE consistently outperforms the other methods on the MNIST (44.30), Fashion MNIST (196.34), and SVHN (77.53) datasets. In machine learning, data scarcity is a common problem, and generative models have the potential to solve it. The variational autoencoder is a generative model that performs variational inference to estimate a low-dimensional posterior dis-tribution given high-dimensional data. Specifically, it optimizes the evidence lower bound from regularization and reconstruction terms, but the two terms are imbalanced in general. If the reconstruction error is not sufficiently small to belong to the population, the generative model performance cannot be guaran-teed. We propose a generative autoencoder (GAE) that uses an autoencoder to first minimize the reconstruction error and then estimate the distribution using latent vectors mapped onto a lower dimension through the encoder. We com-pare the Fréchet inception distances scores of the proposed GAE and nine other variational autoencoders on the MNIST, Fashion MNIST, CIFAR10, and SVHN datasets. The proposed GAE consistently outperforms the other methods on the MNIST (44.30), Fashion MNIST (196.34), and SVHN (77.53) datasets. In machine learning, data scarcity is a common problem, and generative models have the potential to solve it. The variational autoencoder is a generative model that performs variational inference to estimate a low-dimensional posterior dis-tribution given high-dimensional data. Specifically, it optimizes the evidence lower bound from regularization and reconstruction terms, but the two terms are imbalanced in general. If the reconstruction error is not sufficiently small to belong to the population, the generative model performance cannot be guaran-teed. We propose a generative autoencoder (GAE) that uses an autoencoder to first minimize the reconstruction error and then estimate the distribution using latent vectors mapped onto a lower dimension through the encoder. We com-pare the Fréchet inception distances scores of the proposed GAE and nine other variational autoencoders on the MNIST, Fashion MNIST, CIFAR10, and SVHN datasets. The proposed GAE consistently outperforms the other methods on the MNIST (44.30), Fashion MNIST (196.34), and SVHN (77.53) datasets. KCI Citation Count: 0 |
| Author | Kim, YoungSoo Ko, YoungMin Ko, SunWoo |
| Author_xml | – sequence: 1 givenname: YoungMin orcidid: 0000-0003-2779-3170 surname: Ko fullname: Ko, YoungMin organization: Jeonju University – sequence: 2 givenname: SunWoo surname: Ko fullname: Ko, SunWoo organization: Jeonju University – sequence: 3 givenname: YoungSoo surname: Kim fullname: Kim, YoungSoo email: pineland@jj.ac.kr organization: Jeonju University |
| BackLink | https://www.kci.go.kr/kciportal/ci/sereArticleSearch/ciSereArtiView.kci?sereArticleSearchBean.artiId=ART003176255$$DAccess content in National Research Foundation of Korea (NRF) |
| BookMark | eNqFkUFr3DAQhUVJoZu05159LjiRZiRZPoaQpguBQNiexaw8WrRxrSA7W9JfX6-3hVIoPQ0zvG9meO9cnA15YCE-KnmpQbkrnkraX4IErCU25o1YASDWDYI9EysFYGqrLb4T5-O4lxKkNm4lHu544EJTOnBFL1PmIeSOSzXl6rnwgYepygcuhXcvPZX0Y1bmocqxOszd0lD_J_hevI3Uj_zhV70QXz_fbm6-1PcPd-ub6_s6aJRQN8poJ1uF0AZuLKN2GnSg1jVbZZBV58BpJVFZUtRi1EEzRNcxx06TwQvx6bR3KNE_heQzpaXusn8q_vpxs_ZKWmca08zi9UncZdr755K-UXldiGWQy85TmVLo2W8VdqRk1FvSOjpL7WymcggstwH18bA57Qolj2Ph6EOaFiOmQqmfj_pjHH6Jwx_j8Mc4Zu7qL-73H_8m7In4nnp-_Z_c324eQYF1gD8Bdrqg-w |
| CitedBy_id | crossref_primary_10_1016_j_microc_2025_114814 crossref_primary_10_1109_ACCESS_2024_3428479 |
| Cites_doi | 10.48550/arXiv.1606.05908 10.1109/TPAMI.1979.4766926 10.1561/9781680836233 10.48550/arXiv.1701.00160 10.1109/CVPR.2019.01052 10.1007/s10994-019-05791-5 10.24963/ijcai.2017/273 10.1007/978-0-387-45528-0 10.48550/arXiv.1305.1707 10.1002/gamm.202100008 10.48550/arXiv.1611.02731 10.1109/ACCESS.2020.2977671 10.1126/science.290.5500.2323 10.1109/TPAMI.2013.57 10.48550/arXiv.1903.12436 10.48550/arXiv.1903.05789 10.1109/TIP.2020.2964429 10.48550/arXiv.1611.02648 10.1109/TKDE.2006.17 10.48550/arXiv.1312.6114 10.1111/jawr.12182 10.1609/aaai.v33i01.33015885 10.48550/arXiv.1804.00891 10.48550/arXiv.1711.01558 10.1109/TIT.1968.1054102 |
| ContentType | Journal Article |
| Copyright | 1225‐6463/$ © 2024 ETRI |
| Copyright_xml | – notice: 1225‐6463/$ © 2024 ETRI |
| DBID | AAYXX CITATION DOA ACYCR |
| DOI | 10.4218/etrij.2023-0375 |
| DatabaseName | CrossRef DOAJ: Directory of Open Access Journals Korean Citation Index |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | CrossRef |
| Database_xml | – sequence: 1 dbid: DOA name: DOAJ Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering |
| EISSN | 2233-7326 |
| EndPage | 89 |
| ExternalDocumentID | oai_kci_go_kr_ARTI_10685757 oai_doaj_org_article_b13da10f4ba44f86a90371832e0bc345 10_4218_etrij_2023_0375 ETR212682 |
| Genre | article |
| GrantInformation_xml | – fundername: Jeonju University |
| GroupedDBID | -~X .4S .DC 0R~ 29G 2WC 5GY 5VS 9ZL AAKPC AAMMB ACGFS ACXQS ACYCR ADBBV ADDVE ADMLS AEFGJ AENEX AGXDD AIDQK AIDYY ALMA_UNASSIGNED_HOLDINGS ARCSS AVUZU BCNDV DU5 E3Z EBS EDO EJD GROUPED_DOAJ IPNFZ ITG ITH JDI KQ8 KVFHK MK~ ML~ O9- OK1 OVT RIG RNS TR2 TUS WIN XSB AAYXX ALUQN CITATION |
| ID | FETCH-LOGICAL-c4302-71548091329ce76e348424ca987b153e1d828410316a1a93f4c4e2f8deefd4a53 |
| IEDL.DBID | DOA |
| ISICitedReferencesCount | 3 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001201292000001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1225-6463 |
| IngestDate | Sat Oct 25 08:02:08 EDT 2025 Fri Oct 03 12:52:28 EDT 2025 Sat Nov 29 08:11:38 EST 2025 Tue Nov 18 20:55:42 EST 2025 Wed Aug 20 07:26:19 EDT 2025 |
| IsDoiOpenAccess | true |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 1 |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c4302-71548091329ce76e348424ca987b153e1d828410316a1a93f4c4e2f8deefd4a53 |
| Notes | Funding information This study was supported by a research grant of Jeonju University in 2022. https://doi.org/10.4218/etrij.2023-0375 |
| ORCID | 0000-0003-2779-3170 |
| OpenAccessLink | https://doaj.org/article/b13da10f4ba44f86a90371832e0bc345 |
| PageCount | 10 |
| ParticipantIDs | nrf_kci_oai_kci_go_kr_ARTI_10685757 doaj_primary_oai_doaj_org_article_b13da10f4ba44f86a90371832e0bc345 crossref_citationtrail_10_4218_etrij_2023_0375 crossref_primary_10_4218_etrij_2023_0375 wiley_primary_10_4218_etrij_2023_0375_ETR212682 |
| PublicationCentury | 2000 |
| PublicationDate | February 2025 |
| PublicationDateYYYYMMDD | 2025-02-01 |
| PublicationDate_xml | – month: 02 year: 2025 text: February 2025 |
| PublicationDecade | 2020 |
| PublicationTitle | ETRI journal |
| PublicationYear | 2025 |
| Publisher | Electronics and Telecommunications Research Institute (ETRI) 한국전자통신연구원 |
| Publisher_xml | – name: Electronics and Telecommunications Research Institute (ETRI) – name: 한국전자통신연구원 |
| References | 1968; 14 2020; 8 2017; 30 2021; 34 2021; 44 2013; 35 2021 2019; 12 2022; 23 2008 2019 2018 2006 1979; 3 2017 2016 2019; 108 2004 2013 2016; 29 2014; 50 2005; 18 2000; 290 2020; 29 e_1_2_8_28_1 e_1_2_8_29_1 Oord A. (e_1_2_8_34_1) 2017; 30 e_1_2_8_25_1 Cayton L. (e_1_2_8_15_1) 2008 e_1_2_8_26_1 e_1_2_8_27_1 Sønderby C. K. (e_1_2_8_20_1) 2016; 29 e_1_2_8_2_1 e_1_2_8_5_1 e_1_2_8_7_1 Tran L. (e_1_2_8_30_1) 2022; 23 e_1_2_8_6_1 e_1_2_8_9_1 e_1_2_8_8_1 e_1_2_8_21_1 e_1_2_8_22_1 Higgins I. (e_1_2_8_32_1) 2016 Goodfellow I. (e_1_2_8_3_1) 2016 e_1_2_8_17_1 e_1_2_8_18_1 e_1_2_8_39_1 e_1_2_8_19_1 e_1_2_8_36_1 e_1_2_8_14_1 Dai B. (e_1_2_8_24_1) 2021; 34 e_1_2_8_35_1 e_1_2_8_38_1 e_1_2_8_16_1 e_1_2_8_37_1 McLachlan G. J. (e_1_2_8_13_1) 2004 Mohri M. (e_1_2_8_4_1) 2018 Kingma D. P. (e_1_2_8_23_1) 2016; 29 e_1_2_8_10_1 e_1_2_8_31_1 e_1_2_8_11_1 e_1_2_8_12_1 e_1_2_8_33_1 |
| References_xml | – start-page: 1214 year: 2018 end-page: 1223 – start-page: 10275 year: 2019 end-page: 10284 – volume: 44 year: 2021 article-title: An introduction to deep generative modeling publication-title: GAMM‐Mitteilungen – start-page: 2019 5885 end-page: 5892 – volume: 34 start-page: 7180 year: 2021 end-page: 7192 article-title: On the value of infinite gradients in variational autoencoder models publication-title: Adv. Neural Inf. Process Syst. – year: 2021 – volume: 30 year: 2017 article-title: Neural discrete representation learning publication-title: Adv. Neural Inf. Process Syst. – volume: 50 start-page: 1226 year: 2014 end-page: 1241 article-title: Evaluation of CFSR climate data for hydrologic prediction in data‐scarce watersheds: an application in the Blue Nile River Basin publication-title: J. Am. Water Resour. Assoc. – volume: 3 start-page: 306 year: 1979 end-page: 307 article-title: A problem of dimensionality: a simple example publication-title: IEEE Trans. Pattern Anal. Mach. Intell. – volume: 35 start-page: 2765 year: 2013 end-page: 2781 article-title: Sparse subspace clustering: algorithm, theory, and applications publication-title: IEEE Trans. Pattern Anal. Mach. Intell. – volume: 29 year: 2016 article-title: Ladder variational autoencoders publication-title: Adv. Neural Inf. Process Syst. – volume: 290 start-page: 2323 year: 2000 end-page: 2326 article-title: Nonlinear dimensionality reduction by locally linear embedding publication-title: Science – volume: 14 start-page: 55 year: 1968 end-page: 63 article-title: On the mean accuracy of statistical pattern recognizers publication-title: IEEE Trans. Inf. Theory – volume: 29 year: 2016 article-title: Improved variational inference with inverse autoregressive flow publication-title: Adv. Neural Inf. Process Syst. – year: 2016 – year: 2018 – start-page: pp. 1965 year: 2017 end-page: 1972 – volume: 29 start-page: 3665 year: 2020 end-page: 3680 article-title: Zero‐VAE‐GAN: generating unseen features for generalized and transductive zero‐shot learning publication-title: IEEE Trans. Image Process. – volume: 23 start-page: 5010 year: 2022 end-page: 5046 article-title: Cauchy‐Schwarz regularized autoencoder publication-title: J. Mach. Learn. Res. – year: 2008 – start-page: 1558 year: 2016 end-page: 1566 – year: 2006 – year: 2004 – volume: 12 start-page: 307 year: 2019 end-page: 392 – year: 2017 – volume: 18 start-page: 63 year: 2005 end-page: 77 article-title: Training cost‐sensitive neural networks with methods addressing the class imbalance problem publication-title: IEEE Trans. Knowl. Data Eng. – year: 2019 – volume: 8 start-page: 43992 year: 2020 end-page: 44005 article-title: Variational autoencoder with optimizing Gaussian mixture model priors publication-title: IEEE Access – volume: 108 start-page: 1329 year: 2019 end-page: 1351 article-title: Data scarcity, robustness and extreme multi‐label classification publication-title: Mach. Learn. – year: 2013 – ident: e_1_2_8_18_1 doi: 10.48550/arXiv.1606.05908 – ident: e_1_2_8_12_1 doi: 10.1109/TPAMI.1979.4766926 – ident: e_1_2_8_19_1 doi: 10.1561/9781680836233 – ident: e_1_2_8_9_1 doi: 10.48550/arXiv.1701.00160 – ident: e_1_2_8_37_1 – ident: e_1_2_8_39_1 doi: 10.1109/CVPR.2019.01052 – ident: e_1_2_8_5_1 doi: 10.1007/s10994-019-05791-5 – ident: e_1_2_8_26_1 doi: 10.24963/ijcai.2017/273 – volume: 30 year: 2017 ident: e_1_2_8_34_1 article-title: Neural discrete representation learning publication-title: Adv. Neural Inf. Process Syst. – ident: e_1_2_8_2_1 doi: 10.1007/978-0-387-45528-0 – ident: e_1_2_8_7_1 doi: 10.48550/arXiv.1305.1707 – ident: e_1_2_8_10_1 doi: 10.1002/gamm.202100008 – volume: 29 year: 2016 ident: e_1_2_8_20_1 article-title: Ladder variational autoencoders publication-title: Adv. Neural Inf. Process Syst. – volume: 34 start-page: 7180 year: 2021 ident: e_1_2_8_24_1 article-title: On the value of infinite gradients in variational autoencoder models publication-title: Adv. Neural Inf. Process Syst. – ident: e_1_2_8_28_1 – ident: e_1_2_8_21_1 doi: 10.48550/arXiv.1611.02731 – ident: e_1_2_8_31_1 doi: 10.1109/ACCESS.2020.2977671 – volume: 23 start-page: 5010 year: 2022 ident: e_1_2_8_30_1 article-title: Cauchy‐Schwarz regularized autoencoder publication-title: J. Mach. Learn. Res. – volume-title: Foundations of machine learning year: 2018 ident: e_1_2_8_4_1 – ident: e_1_2_8_16_1 doi: 10.1126/science.290.5500.2323 – volume-title: beta‐VAE: learning basic visual concepts with a constrained variational framework year: 2016 ident: e_1_2_8_32_1 – ident: e_1_2_8_17_1 doi: 10.1109/TPAMI.2013.57 – ident: e_1_2_8_36_1 doi: 10.48550/arXiv.1903.12436 – ident: e_1_2_8_22_1 doi: 10.48550/arXiv.1903.05789 – ident: e_1_2_8_38_1 doi: 10.1109/TIP.2020.2964429 – ident: e_1_2_8_27_1 doi: 10.48550/arXiv.1611.02648 – ident: e_1_2_8_8_1 doi: 10.1109/TKDE.2006.17 – ident: e_1_2_8_14_1 doi: 10.48550/arXiv.1312.6114 – volume: 29 year: 2016 ident: e_1_2_8_23_1 article-title: Improved variational inference with inverse autoregressive flow publication-title: Adv. Neural Inf. Process Syst. – ident: e_1_2_8_6_1 doi: 10.1111/jawr.12182 – ident: e_1_2_8_25_1 doi: 10.1609/aaai.v33i01.33015885 – ident: e_1_2_8_35_1 doi: 10.48550/arXiv.1804.00891 – volume-title: Algorithms for manifold learning, eScholarship year: 2008 ident: e_1_2_8_15_1 – ident: e_1_2_8_33_1 doi: 10.48550/arXiv.1711.01558 – volume-title: Discriminant analysis and statistical pattern recognition year: 2004 ident: e_1_2_8_13_1 – ident: e_1_2_8_29_1 – ident: e_1_2_8_11_1 doi: 10.1109/TIT.1968.1054102 – volume-title: Deep learning year: 2016 ident: e_1_2_8_3_1 |
| SSID | ssj0020458 |
| Score | 2.3799887 |
| Snippet | In machine learning, data scarcity is a common problem, and generative models have the potential to solve it. The variational autoencoder is a generative model... |
| SourceID | nrf doaj crossref wiley |
| SourceType | Open Website Enrichment Source Index Database Publisher |
| StartPage | 80 |
| SubjectTerms | autoencoder data augmentation dimensionality reduction generative model variational autoencoder 전자/정보통신공학 |
| SummonAdditionalLinks | – databaseName: Wiley Online Library Free Content dbid: WIN link: http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1LS8NAEF6keNCDb7G-COjBS2oek9dRpWJBqkjV3pZNsltqJZE19vc7s4mlFUQETyFhJxlmd2ZnNrvfx9ipK9NAOQFWJ27u2yBTxxZxJG1M5kKBKUKgzKn3p9uo34-Hw-S-2U1IZ2FqfIjZght5honX5OAiNSwkgLMSdWKlxy8dYv-2iccVo7ALxjWfe_1ZyUW_AankwlFrhxD6NbgPveH8m_zCvGTg-3G2KbRaTFrNrHO9_g_6brC1JuW0LuoxssmWZLHFVueACLfZXY0-TaHPEh9VSeiWudRWVVpvNcaTRVs9tSGu183RTatU1hTvmuXEecEd9njdHVzd2A3Vgp2BjzExosqFIEK9JJNRKH2IwYNMJHGUYkyUbo6VGRAlRChckfgKMpCeinMpVQ4i8HdZqygLuccspQI3lBE18jA6YHtHUZmVQAoKEqfNOl-G5lmDQ050GK8c6xGyFTeW4mQpTpZqs7OZwFsNwfFz00vquVkzws42D0o94o0r8tT1c6NUKgAUKpgQbCFGNumkmQ_4khPsdz7JxkaerqOSTzTHCqOHXw6J1DRqs3PT27-pxLuDB0wPwtjb_7PEAVvxiG7YbBI_ZK1Kf8gjtpxNq_G7PjYj_RMepv8M priority: 102 providerName: Wiley-Blackwell |
| Title | Generative autoencoder to prevent overregularization of variational autoencoder |
| URI | https://onlinelibrary.wiley.com/doi/abs/10.4218%2Fetrij.2023-0375 https://doaj.org/article/b13da10f4ba44f86a90371832e0bc345 https://www.kci.go.kr/kciportal/ci/sereArticleSearch/ciSereArtiView.kci?sereArticleSearchBean.artiId=ART003176255 |
| Volume | 47 |
| WOSCitedRecordID | wos001201292000001&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 | |
| ispartofPNX | ETRI Journal, 2025, 47(1), , pp.80-89 |
| journalDatabaseRights | – providerCode: PRVAON databaseName: DOAJ Directory of Open Access Journals customDbUrl: eissn: 2233-7326 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0020458 issn: 1225-6463 databaseCode: DOA dateStart: 20170101 isFulltext: true titleUrlDefault: https://www.doaj.org/ providerName: Directory of Open Access Journals – providerCode: PRVWIB databaseName: Wiley Online Library Free Content customDbUrl: eissn: 2233-7326 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0020458 issn: 1225-6463 databaseCode: WIN dateStart: 19970101 isFulltext: true titleUrlDefault: https://onlinelibrary.wiley.com providerName: Wiley-Blackwell |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1NT4NAEN2YxoMejJ-xfoVED15o-RgWOKppYxNTjana22aBXVNrSoO0v9-ZhTbtwfTiBQLZDZvHsDMPlvcYu3FVEmgnQHbiZr4NKnFsGYXKxmKOSywRAm3-en9_Cvv9aDiMX1asvmhNWCUPXAHXTlw_k66jIZEAOuIyJpE5jEPlJKkPRr3UCeMFmaqpFn3-I6qF0Wpz4H4l6gOYz9pkVPXVItdwm_xf1_KRke3HLDMp9HqxarJNd5_t1WWidVcN74Btqckh210RDzxiz5ViNE1XlpyVOSlSZqqwytyaVrpMFi3PLIzZfFH_bmnl2prjUf0KcLXjMXvrdgYPj3Ztj2Cn4OM8FhLbIFlPL05VyJUPEXiQyjgKE5zHlJshmwKyceDSlbGvIQXl6ShTSmcgA_-ENSb5RJ0yS-vA5SqkRh4-0dwAjpVRDAloiJ0may1AEmmtHU4WFt8COQShKgyqglAVhGqT3S47TCvZjL-b3hPqy2akd21OYBSIOgrEpihosmu8Z2Kcjkx_2n_mYlwIZAU9vDInI9Kwydrmnm4akugMXjGl88g7-4_BnbMdj1yDzVrvC9Yoi5m6ZNvpvBz9FFcmcHH70ev_Ap4I7rU |
| linkProvider | Directory of Open Access Journals |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV3dT9swED8NmMT2AIwPrYONSOyBl0A-Lk7yOBBVK0qRUAd9s_Jho8LUIC_079-dE6qChBDSnqJEtnO6853vnPj3A_jpqzzSXkTViV-GLqrcc7MkVi4lcyKjFCHS9tT79SAeDpPxOF08C9PgQ8w33NgzbLxmB-cNafZypGWJrVibyd0R03-7TOS6BCsoMGUWg5v-cF508YdALrpo3roCRdjA-_AQxy8GeLYyWQB_Wm-mRj9PW-26013_HxJvwFqbdTq_mmnyBT6o6SZ8XsAi3ILLBoCao5-TPdYVA1yWyjh15Tw0ME8O_-1pLHe9aU9vOpV2ZnTX7igudtyG392z0WnPbdkW3AJDCosxFy-MEhqkhYqFCjHBAIssTeKcwqLySyrOkFkhROZnaaixQBXopFRKl5hF4Q4sT6up-gqO1pEvVMyNAgoQ1N7TXGmlmKPG1OvA0ZOmZdFCkTMjxh9JJQnrSlpNSdaUZE114HDe4aFB4Xi96Qmbbt6M4bPtg8rcytYbZe6HpRUqzxA1CZgyciEFN-XlRYg0yAEZXt4XE9ufr7eVvDeSiow-vVkwr2ncgWNr7rdEkmejK8oQRBJ8e3ePfVjtjS4GctAfnu_Cp4DZh-0_43uwXJtH9R0-FrN68tf8sNP-HzZAA0Q |
| linkToPdf | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV3dT9tADLegIMQexudEN9gisQdeUvLhfD0OaLVqqCDUQd9O-bhDHaipjtC_f_YlVAUJISSeokT2xbLPPju5-xngpyuzQDkBVSdu4dsoM8dO40jalMyFKaUIgTKn3q_Po8EgHo2SxbMwNT7E_IMbe4aJ1-zgcloo9nKkZYmtWOnxvw63_7a5kesyrCBl47yt66Y_mBdd_COQiy6at3aIoV_D-_AQxy8GeLYyGQB_Wm8mWj1PW82609v4CIk34XOTdVq_6mmyBUtysg2fFrAId-CiBqDm6Gelj1XJAJeF1FZVWtMa5sni3Z7a9K7XzelNq1TWjO6aL4qLjLvwt9cdnv62m24Ldo4-hcWIixdGCfWSXEah9DFGD_M0iaOMwqJ0CyrOkLtChKmbJr7CHKWn4kJKVWAa-F-gNSkncg8spQI3lBETeRQgiN5RXGklmKHCxGlD50nTIm-gyLkjxr2gkoR1JYymBGtKsKbacDRnmNYoHK-TnrDp5mQMn20elPpWNN4oMtcvjFBZiqhIwISRCym4SSfLfaRBDsnw4i4fG36-3pbiTgsqMvr05pD7mkZtODbmfksk0R1eUYYQxt7Xd3P8gLXLs5447w_-fIN1j5sPmy3j-9Cq9KM8gNV8Vo0f9Hcz6_8Du50Cvw |
| 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=Generative+autoencoder+to+prevent+overregularization+of+variational+autoencoder&rft.jtitle=ETRI+journal&rft.au=YoungMin+Ko&rft.au=SunWoo+Ko&rft.au=YoungSoo+Kim&rft.date=2025-02-01&rft.pub=Electronics+and+Telecommunications+Research+Institute+%28ETRI%29&rft.issn=1225-6463&rft.eissn=2233-7326&rft.volume=47&rft.issue=1&rft.spage=80&rft.epage=89&rft_id=info:doi/10.4218%2Fetrij.2023-0375&rft.externalDBID=DOA&rft.externalDocID=oai_doaj_org_article_b13da10f4ba44f86a90371832e0bc345 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1225-6463&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1225-6463&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1225-6463&client=summon |