Enhancing Dimensionality Reduction in Driving Behavior Learning: Integrating SENet with VAE

This study addresses a common limitation of conventional Variational Autoencoder (VAE)-based methods in dimensionality reduction for state representation learning, especially in autonomous driving, by integrating Squeeze-and-Excitation Networks (SENet) into the VAE framework. While traditional VAE a...

Full description

Saved in:
Bibliographic Details
Published inInternational Journal of Networking and Computing Vol. 15; no. 2; pp. 138 - 152
Main Authors Uehara, Yuta, Matsumae, Susumu
Format Journal Article
LanguageEnglish
Published IJNC Editorial Committee 2025
Subjects
Online AccessGet full text
ISSN2185-2839
2185-2847
2185-2847
DOI10.15803/ijnc.15.2_138

Cover

Abstract This study addresses a common limitation of conventional Variational Autoencoder (VAE)-based methods in dimensionality reduction for state representation learning, especially in autonomous driving, by integrating Squeeze-and-Excitation Networks (SENet) into the VAE framework. While traditional VAE approaches effectively handle high-dimensional data with reduced computational costs, they often struggle to adequately capture complex features in certain tasks. To overcome this challenge, we propose the SENet-VAE model, which incorporates SENet into the VAE architecture, and evaluate its performance in driving behavior learning using deep reinforcement learning. Our experiments compare three setups: raw image data, conventional VAE, and SENet-VAE. Furthermore, we examine how the placement and number of SE-Blocks affect performance. The results demonstrate that SENet-VAE surpasses the limitations of conventional VAE and achieves superior accuracy in learning. This work highlights the potential of SENet-VAE as a robust dimensionality reduction solution for state representation learning.
AbstractList This study addresses a common limitation of conventional Variational Autoencoder (VAE)-based methods in dimensionality reduction for state representation learning, especially in autonomous driving, by integrating Squeeze-and-Excitation Networks (SENet) into the VAE framework. While traditional VAE approaches effectively handle high-dimensional data with reduced computational costs, they often struggle to adequately capture complex features in certain tasks. To overcome this challenge, we propose the SENet-VAE model, which incorporates SENet into the VAE architecture, and evaluate its performance in driving behavior learning using deep reinforcement learning. Our experiments compare three setups: raw image data, conventional VAE, and SENet-VAE. Furthermore, we examine how the placement and number of SE-Blocks affect performance. The results demonstrate that SENet-VAE surpasses the limitations of conventional VAE and achieves superior accuracy in learning. This work highlights the potential of SENet-VAE as a robust dimensionality reduction solution for state representation learning.
Author Yuta Uehara
Susumu Matsumae
Author_xml – sequence: 1
  givenname: Yuta
  surname: Uehara
  fullname: Uehara, Yuta
– sequence: 2
  givenname: Susumu
  surname: Matsumae
  fullname: Matsumae, Susumu
BookMark eNqFkE9vwjAMxaOJSWOM6879AoWkJWqywyQG3YaENmn_LjtUbnBoUElRGkB8-7Uwod3mi5_t9_PhXZOOrSwScsvogHFB46FZWdXIQZSxWFyQbsQEDyMxSjpnHcsr0q_rFW0qSSSN4i75Tm0BVhm7DKZmjbY2lYXS-EPwhout8s0YGBtMndm1ngcsYGcqF8wRnG02d8HMelw68O35PX1BH-yNL4KvcXpDLjWUNfZ_e498PqYfk-dw_vo0m4znoYooFeFCwwI5l8hBRAo1MKW1kjDSXHPOE5aDSmKUMhEyj6hWWjHJaJ4IofOGjXtkePq7tRs47KEss40za3CHjNHsGE_WxtPI7BhPQwxOhHJVXTvU_wP3J2BVe1ji2Q7OG1XiXzdvgfNBFeAytPEPI9KC8Q
Cites_doi 10.1007/s10596-022-10159-1
10.1109/CVPR.2018.00745
10.1109/CANDARW64572.2024.00069
10.1109/ICRA40945.2020.9197465
10.1109/ICRA.2019.8793742
10.1109/CANDARW60564.2023.00064
10.3389/fnbot.2023.1301785
10.3390/s20215991
ContentType Journal Article
Copyright 2025 International Journal of Networking and Computing
Copyright_xml – notice: 2025 International Journal of Networking and Computing
DBID AAYXX
CITATION
ADTOC
UNPAY
DOI 10.15803/ijnc.15.2_138
DatabaseName CrossRef
Unpaywall for CDI: Periodical Content
Unpaywall
DatabaseTitle CrossRef
DatabaseTitleList
Database_xml – sequence: 1
  dbid: UNPAY
  name: Unpaywall
  url: https://proxy.k.utb.cz/login?url=https://unpaywall.org/
  sourceTypes: Open Access Repository
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISSN 2185-2847
EndPage 152
ExternalDocumentID 10.15803/ijnc.15.2_138
10_15803_ijnc_15_2_138
article_ijnc_15_2_15_138_article_char_en
GroupedDBID 7.U
ALMA_UNASSIGNED_HOLDINGS
JSF
JSH
KQ8
KWQ
OK1
RJT
RZJ
AAYXX
CITATION
ISHAI
ADTOC
UNPAY
ID FETCH-LOGICAL-c2008-dfade559e5a82cefa1cffc9a4f5f55571bac73e99789b20fcfc1910b788fbfad3
IEDL.DBID UNPAY
ISSN 2185-2839
2185-2847
IngestDate Sun Sep 07 10:48:35 EDT 2025
Wed Oct 01 05:46:01 EDT 2025
Wed Sep 03 06:30:38 EDT 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 2
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c2008-dfade559e5a82cefa1cffc9a4f5f55571bac73e99789b20fcfc1910b788fbfad3
OpenAccessLink https://proxy.k.utb.cz/login?url=https://doi.org/10.15803/ijnc.15.2_138
PageCount 15
ParticipantIDs unpaywall_primary_10_15803_ijnc_15_2_138
crossref_primary_10_15803_ijnc_15_2_138
jstage_primary_article_ijnc_15_2_15_138_article_char_en
ProviderPackageCode CITATION
AAYXX
PublicationCentury 2000
PublicationDate 2025
2025-00-00
PublicationDateYYYYMMDD 2025-01-01
PublicationDate_xml – year: 2025
  text: 2025
PublicationDecade 2020
PublicationTitle International Journal of Networking and Computing
PublicationTitleAlternate IJNC
PublicationYear 2025
Publisher IJNC Editorial Committee
Publisher_xml – name: IJNC Editorial Committee
References [12] S. Song, F. Yu, X. Jiang, J. Zhu, W. Cheng, and X. Fang. Loop closure detection of visual SLAM based on variationalautoencoder. Frontiers in Neurorobotics, 17, 2024. doi: 10.3389/fnbot.2023.1301785.
[5] A. Kendall, J. Hawke, D. Janz, P. Mazur, D. Reda, J.-M. Allen, V.-D. Lam,A. Bewley, and A. Shah. Learning to drive in a day. In 2019 International Conference on Robotics and Automation(ICRA), pages 8248–8254, 2019.
[7] Y. Uehara and S. Matsumae. Dimensionality reduction methods using VAE for deep reinforcementlearning of autonomous driving. In International Workshop on Advances in Networking andComputing (WANC), CANDARW, 2023.
[13] T. Zhang, Y. Yang, and A. Zhang. 3D reconstruction of porous media using a batch normalizedvariational auto-encoder. Computational Geosciences, 26:1261–1278, 2022. doi: 10.1007/s10596-022-10159-1.
[6] A. Gupta, A. S. Khwaja, A. Anpalagan, L. Guan, and B. Venkatesh. Policy-gradient and actor-critic based state representation learningfor safe driving of autonomous vehicles. Sensors, 20(21):5991, 2020.
[14] T. Kramer. OpenAI Gym environments for Donkey Car [source code]. https://github.com/tawnkramer/gym-donkeycar.git, 2018.
[8] J. Schulman, F. Wolski, P. Dhariwal, A. Radford, and O. Klimov. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347, 2017.
[9] J. Hu, L. Shen, and G. Sun. Squeeze-and-excitation networks. In Proceedings of the IEEE Conference on Computer Vision andPattern Recognition (CVPR), 2018.
[15] A. Raffin, A. Hill, A. Gleave, A. Kanervisto, M. Ernestus, and N. Dormann. Stable-baselines3: Reliable reinforcement learning implementations. The Journal of Machine Learning Research, 22(1):12348–12355,2021.
[3] B. Balaji, S. Mallya, S. Genc, S. Gupta, L. Dirac, et al. DeepRacer: Autonomous racing platform for experimentation with Sim2Real reinforcement learning. In 2020 IEEE International Conference on Robotics and Automation (ICRA), pages 2746–2754, Paris, France, 2020. doi: 10.1109/ICRA40945.2020.9197465.
[1] Q. Zhang, T. Du, and C. Tian. Self-driving scale car trained by deep reinforcement learning. arXiv preprint arXiv:1909.03467, 2019.
[10] Y. Uehara and S. Matsumae. Effect of integrating variational autoencoder with SENet ondimensionality reduction in driving behavior learning. In International Workshop on Advances in Networking andComputing (WANC), CANDARW, 2024.
[2] S. Wang, D. Jia, and X. Weng. Deep reinforcement learning for autonomous driving. arXiv preprint arXiv:1811.11329, 2018.
[11] Donkey Car – Home. https://www.donkeycar.com/.
[4] D. P. Kingma and M. Welling. Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114, 2013.
11
12
13
14
15
1
2
3
4
5
6
7
8
9
10
References_xml – reference: [3] B. Balaji, S. Mallya, S. Genc, S. Gupta, L. Dirac, et al. DeepRacer: Autonomous racing platform for experimentation with Sim2Real reinforcement learning. In 2020 IEEE International Conference on Robotics and Automation (ICRA), pages 2746–2754, Paris, France, 2020. doi: 10.1109/ICRA40945.2020.9197465.
– reference: [9] J. Hu, L. Shen, and G. Sun. Squeeze-and-excitation networks. In Proceedings of the IEEE Conference on Computer Vision andPattern Recognition (CVPR), 2018.
– reference: [15] A. Raffin, A. Hill, A. Gleave, A. Kanervisto, M. Ernestus, and N. Dormann. Stable-baselines3: Reliable reinforcement learning implementations. The Journal of Machine Learning Research, 22(1):12348–12355,2021.
– reference: [10] Y. Uehara and S. Matsumae. Effect of integrating variational autoencoder with SENet ondimensionality reduction in driving behavior learning. In International Workshop on Advances in Networking andComputing (WANC), CANDARW, 2024.
– reference: [7] Y. Uehara and S. Matsumae. Dimensionality reduction methods using VAE for deep reinforcementlearning of autonomous driving. In International Workshop on Advances in Networking andComputing (WANC), CANDARW, 2023.
– reference: [8] J. Schulman, F. Wolski, P. Dhariwal, A. Radford, and O. Klimov. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347, 2017.
– reference: [4] D. P. Kingma and M. Welling. Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114, 2013.
– reference: [2] S. Wang, D. Jia, and X. Weng. Deep reinforcement learning for autonomous driving. arXiv preprint arXiv:1811.11329, 2018.
– reference: [6] A. Gupta, A. S. Khwaja, A. Anpalagan, L. Guan, and B. Venkatesh. Policy-gradient and actor-critic based state representation learningfor safe driving of autonomous vehicles. Sensors, 20(21):5991, 2020.
– reference: [13] T. Zhang, Y. Yang, and A. Zhang. 3D reconstruction of porous media using a batch normalizedvariational auto-encoder. Computational Geosciences, 26:1261–1278, 2022. doi: 10.1007/s10596-022-10159-1.
– reference: [5] A. Kendall, J. Hawke, D. Janz, P. Mazur, D. Reda, J.-M. Allen, V.-D. Lam,A. Bewley, and A. Shah. Learning to drive in a day. In 2019 International Conference on Robotics and Automation(ICRA), pages 8248–8254, 2019.
– reference: [11] Donkey Car – Home. https://www.donkeycar.com/.
– reference: [14] T. Kramer. OpenAI Gym environments for Donkey Car [source code]. https://github.com/tawnkramer/gym-donkeycar.git, 2018.
– reference: [12] S. Song, F. Yu, X. Jiang, J. Zhu, W. Cheng, and X. Fang. Loop closure detection of visual SLAM based on variationalautoencoder. Frontiers in Neurorobotics, 17, 2024. doi: 10.3389/fnbot.2023.1301785.
– reference: [1] Q. Zhang, T. Du, and C. Tian. Self-driving scale car trained by deep reinforcement learning. arXiv preprint arXiv:1909.03467, 2019.
– ident: 2
– ident: 13
  doi: 10.1007/s10596-022-10159-1
– ident: 4
– ident: 1
– ident: 11
– ident: 9
  doi: 10.1109/CVPR.2018.00745
– ident: 10
  doi: 10.1109/CANDARW64572.2024.00069
– ident: 3
  doi: 10.1109/ICRA40945.2020.9197465
– ident: 14
– ident: 15
– ident: 5
  doi: 10.1109/ICRA.2019.8793742
– ident: 7
  doi: 10.1109/CANDARW60564.2023.00064
– ident: 12
  doi: 10.3389/fnbot.2023.1301785
– ident: 8
– ident: 6
  doi: 10.3390/s20215991
SSID ssj0000779023
Score 2.2809634
Snippet This study addresses a common limitation of conventional Variational Autoencoder (VAE)-based methods in dimensionality reduction for state representation...
SourceID unpaywall
crossref
jstage
SourceType Open Access Repository
Index Database
Publisher
StartPage 138
SubjectTerms Autonomous driving
Deep reinforcement learning
Dimensionality reduction
Squeeze-and-excitation network
Variational autoencoder
Title Enhancing Dimensionality Reduction in Driving Behavior Learning: Integrating SENet with VAE
URI https://www.jstage.jst.go.jp/article/ijnc/15/2/15_138/_article/-char/en
https://doi.org/10.15803/ijnc.15.2_138
UnpaywallVersion publishedVersion
Volume 15
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
ispartofPNX International Journal of Networking and Computing, 2025, Vol.15(2), pp.138-152
journalDatabaseRights – providerCode: PRVAFT
  databaseName: Open Access Digital Library
  customDbUrl:
  eissn: 2185-2847
  dateEnd: 99991231
  omitProxy: true
  ssIdentifier: ssj0000779023
  issn: 2185-2847
  databaseCode: KQ8
  dateStart: 20110101
  isFulltext: true
  titleUrlDefault: http://grweb.coalliance.org/oadl/oadl.html
  providerName: Colorado Alliance of Research Libraries
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV3dT8IwEL8ImPjktxGjpg8m-jIc29ptvhEFUSPxC4PxYWlLiyCZhkCM_vX2WCGoD_jWpLe2ufbS-3V3vwM4oAyBtGSOFwnmBAwfmhgLnNCTTPsGFHnjIJrrBqs3g8sWbdkAWcyFmf1_TyPXP-72jJ2UaclLyn6Ug4IZjbl5KDQbN5UnrBxnLhwz0bhkmG0HoWVn_DvAj9tnsWccMMysXxql7_zzg_f7M3dLbQXOJ6vKQkpeS6OhKMmvX4SN85e9CsvWvSSV7DyswYJK12FlUrqBWEvegOdq-oJMG2mHnCG_f8bNYTxycodcrrhbpJuSs0EXHxyIZVEcEEvH2jkhF5ZnArvvqw01JPikSx4r1U1o1qoPp3XH1llw5Dj6oa15WxlkoSiPPKk0L0utZcwDTTWlNCwLLkNfxQZwxsJztdTSoDxXGPSshfnW34J8-paqbSBt5TLOYh-9skBwzY0LxyMeqEjEEXNFEQ4n-k_eMzqNBGEI6ixBnZlmMtZZEcJse6Zy1pRmxShKTjswT80YexGOpvs5Z5Kd_4vuQn44GKk944IMxT7krm6jfXsKvwEo19tf
linkProvider Unpaywall
linkToUnpaywall http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV3dS8MwEA-6CT45P3GikgdBXzr7lbT1bbjOKThEnUx8KEmWzM1Rx-gQ_evNtdmY-jDfArkm4ZIj90vvfofQCaEApAW13JBTy6fw0ESpbwWuoMrToMjNg2hu27TV8W-6pGsCZCEXZvH_PQlt73ww1HbikJqbOF64isp6NGqXULnTvqs_Q-U4feHoifKSYabtB4ad8e8AP26ftaF2wCCzfn2ajtnnBxuNFu6WZgVdzVZVhJS81aYZr4mvX4SNy5e9iTaMe4nrxXnYQisy3UaVWekGbCx5B73E6SswbaR93AB-_4KbQ3vk-B64XGG38CDFjckAHhywYVGcYEPH2r_A14ZnArof4rbMMDzp4qd6vIs6zfjxsmWZOguWyKMfeor1pEYWkrDQFVIxRyglIuYrogghgcOZCDwZacAZcddWQgmN8myu0bPi-ltvD5XS91TuI9yTNmU08sAr8zlTTLtwLGS-DHkUUptX0elM_8m4oNNIAIaAzhLQmW4muc6qKCi2Zy5nTGlRjIDkvAPy1LSxV9HZfD-XTHLwf9FDVMomU3mkXZCMH5vz9w0KDdpq
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=Enhancing+Dimensionality+Reduction+in+Driving+Behavior+Learning%3A+Integrating+SENet+with+VAE&rft.jtitle=International+Journal+of+Networking+and+Computing&rft.au=Susumu+Matsumae&rft.au=Yuta+Uehara&rft.date=2025&rft.pub=IJNC+Editorial+Committee&rft.issn=2185-2839&rft.eissn=2185-2847&rft.volume=15&rft.issue=2&rft.spage=138&rft.epage=152&rft_id=info:doi/10.15803%2Fijnc.15.2_138&rft.externalDocID=article_ijnc_15_2_15_138_article_char_en
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2185-2839&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2185-2839&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2185-2839&client=summon