A convolutional autoencoder-based method for learning and ranking personal daylighting preference
•Novel method for leaning personal binary daylighting preference.•Two-stage training method: feature extraction module and preference inference module.•CAE-based feature extractor designed to monitor unseen luminance map characteristics.•Inference module: classifying daylighting preference and estim...
Saved in:
| Published in: | Building and environment Vol. 285; p. 113595 |
|---|---|
| Main Authors: | , |
| Format: | Journal Article |
| Language: | English |
| Published: |
Elsevier Ltd
01.11.2025
|
| Subjects: | |
| ISSN: | 0360-1323 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | •Novel method for leaning personal binary daylighting preference.•Two-stage training method: feature extraction module and preference inference module.•CAE-based feature extractor designed to monitor unseen luminance map characteristics.•Inference module: classifying daylighting preference and estimating ordinal ranking.
This paper presents a two-stage training method for inferring the relative daylight preference from pairs of luminance maps using convolutional autoencoder (CAE) and relative ranking concepts. It combines an updatable CAE-based feature extraction module and a binary daylighting preference inference module including relative ranking inference. A Python-based target area selection program was developed to enable the CAE model to compress the window and background areas separately from the input luminance map, ensuring that the compressed latent pixels represent these two areas distinctly. The developed CAE model was trained first using luminance maps collected from 11 individuals in private offices. Then, the trained CAE model's encoder was transferred to be a feature extractor of the personal visual preference learning models. Each model was trained to classify binary preference feedback between pairs of visual scenes and infer ordinal ranking scores for luminance distribution characteristics. The results showed that personal daylighting preference could be classified with over 85 % accuracy, and the model successfully identified the most preferred luminance scene by comparing ranking scores. In addition, the trained CAE model was able to recognize luminance maps containing significantly different characteristics by monitoring the reconstruction performance. Compared to a convolutional neural network (CNN)-based approach, the CAE-based model could leverage the condensed pixels and compare real-time luminance map characteristics with the selected conditions using L2 norm and Euclidean distance metrics, which can guide visual adjustments towards the most preferred daylighting conditions. Therefore, this study presents a significant step towards preference-based daylighting control. |
|---|---|
| AbstractList | •Novel method for leaning personal binary daylighting preference.•Two-stage training method: feature extraction module and preference inference module.•CAE-based feature extractor designed to monitor unseen luminance map characteristics.•Inference module: classifying daylighting preference and estimating ordinal ranking.
This paper presents a two-stage training method for inferring the relative daylight preference from pairs of luminance maps using convolutional autoencoder (CAE) and relative ranking concepts. It combines an updatable CAE-based feature extraction module and a binary daylighting preference inference module including relative ranking inference. A Python-based target area selection program was developed to enable the CAE model to compress the window and background areas separately from the input luminance map, ensuring that the compressed latent pixels represent these two areas distinctly. The developed CAE model was trained first using luminance maps collected from 11 individuals in private offices. Then, the trained CAE model's encoder was transferred to be a feature extractor of the personal visual preference learning models. Each model was trained to classify binary preference feedback between pairs of visual scenes and infer ordinal ranking scores for luminance distribution characteristics. The results showed that personal daylighting preference could be classified with over 85 % accuracy, and the model successfully identified the most preferred luminance scene by comparing ranking scores. In addition, the trained CAE model was able to recognize luminance maps containing significantly different characteristics by monitoring the reconstruction performance. Compared to a convolutional neural network (CNN)-based approach, the CAE-based model could leverage the condensed pixels and compare real-time luminance map characteristics with the selected conditions using L2 norm and Euclidean distance metrics, which can guide visual adjustments towards the most preferred daylighting conditions. Therefore, this study presents a significant step towards preference-based daylighting control. |
| ArticleNumber | 113595 |
| Author | Tzempelikos, Athanasios Mah, Dongjun |
| Author_xml | – sequence: 1 givenname: Dongjun orcidid: 0000-0002-5181-2944 surname: Mah fullname: Mah, Dongjun email: dmah@purdue.edu organization: Lyles School of Civil and Construction Engineering, Purdue University, 550 Stadium Mall Dr., West Lafayette, IN 47907, USA – sequence: 2 givenname: Athanasios orcidid: 0000-0001-6788-8372 surname: Tzempelikos fullname: Tzempelikos, Athanasios organization: Lyles School of Civil and Construction Engineering, Purdue University, 550 Stadium Mall Dr., West Lafayette, IN 47907, USA |
| BookMark | eNqFkM1uwjAQhH2gUoH2Faq8QNK1HSf4VoT6JyH10p4tx96AabCRHZB4-ybQnnva1WpmdvTNyMQHj4Q8UCgo0OpxVzRH11n0p4IBEwWlXEgxIVPgFeSUM35LZintYBBLXk6JXmYm-FPojr0LXneZPvYBvQkWY97ohDbbY78NNmtDzDrU0Tu_ybS3WdT-e9wPGNPFavW5c5ttfzlGbDEOQXhHblrdJbz_nXPy9fL8uXrL1x-v76vlOjeson1etdDIUojaSESQ0jKorWElWDQLgEZwy4FZUTJqgJm6NXVTCaElb5tFZVo-J9U118SQ0vBeHaLb63hWFNQIR-3UHxw1wlFXOIPx6WrEod3JYVTJuLG5dRFNr2xw_0X8AI59d18 |
| Cites_doi | 10.1016/j.enbuild.2022.111893 10.1016/j.autcon.2021.103812 10.1016/j.enbuild.2020.110228 10.1016/j.buildenv.2020.107212 10.1016/j.enbuild.2014.03.035 10.1016/j.enbuild.2020.110026 10.1016/j.buildenv.2021.108066 10.1177/14771535221117365 10.1016/j.enbuild.2019.03.046 10.1016/j.buildenv.2016.09.009 10.1080/09613210701544061 10.1016/j.buildenv.2016.08.005 10.1016/j.buildenv.2012.09.017 10.1016/j.enbuild.2013.04.020 10.1016/j.enbuild.2024.114130 10.1016/j.enbuild.2023.113738 10.1016/j.enbuild.2023.112876 10.1016/j.enbuild.2023.113876 10.1191/1365782804li115oa 10.1016/j.enbuild.2018.01.024 10.1177/136578280603800210 10.1016/j.enbuild.2013.11.053 10.1016/j.autcon.2024.105537 10.1016/j.buildenv.2021.107623 10.1016/j.buildenv.2020.107013 10.3390/jimaging4010019 10.1016/j.buildenv.2016.09.018 10.1016/j.applthermaleng.2023.121545 10.1016/j.buildenv.2022.108857 10.1016/j.buildenv.2019.106538 10.1016/j.autcon.2020.103370 10.1016/j.enbuild.2017.02.058 10.1016/j.enbuild.2006.03.017 10.1080/15502724.2014.881720 10.1016/j.buildenv.2021.108498 10.1016/j.enbuild.2023.113051 10.1016/j.buildenv.2020.107397 10.1016/j.apenergy.2017.12.005 10.1080/15502724.2015.1062392 10.1016/j.enbuild.2024.114071 10.1016/j.scs.2018.12.025 10.1582/LEUKOS.2010.07.02003 10.3390/buildings9100219 10.1016/j.buildenv.2016.11.024 10.1016/j.scs.2022.104007 10.1016/j.buildenv.2024.112128 10.1007/s11263-015-0816-y 10.1016/j.enbuild.2017.11.054 10.1080/15502724.2019.1684319 10.1016/j.enbuild.2024.114575 10.1016/j.buildenv.2021.108346 10.1016/j.buildenv.2018.04.022 10.1177/1477153512458671 10.1582/LEUKOS.2007.04.02.002 10.1016/j.enbuild.2018.08.010 10.3368/le.86.3.530 |
| ContentType | Journal Article |
| Copyright | 2025 |
| Copyright_xml | – notice: 2025 |
| DBID | AAYXX CITATION |
| DOI | 10.1016/j.buildenv.2025.113595 |
| DatabaseName | CrossRef |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering |
| ExternalDocumentID | 10_1016_j_buildenv_2025_113595 S0360132325010674 |
| GroupedDBID | --K --M -~X .~1 0R~ 1B1 1RT 1~. 1~5 23N 4.4 457 4G. 5GY 5VS 7-5 71M 8P~ 9JM 9JN AABNK AAEDT AAEDW AAHCO AAIKC AAIKJ AAKOC AALRI AAMNW AAOAW AAQFI AARJD AATTM AAXKI AAXUO AAYWO ABFNM ABFYP ABJNI ABLST ABMAC ACDAQ ACGFS ACIWK ACLOT ACRLP ACVFH ADBBV ADCNI ADEZE ADTZH AEBSH AECPX AEIPS AEKER AENEX AEUPX AFJKZ AFPUW AFRAH AFTJW AFXIZ AGHFR AGUBO AGYEJ AHEUO AHHHB AHIDL AHJVU AIEXJ AIGII AIIUN AIKHN AITUG AKBMS AKIFW AKRWK AKYEP ALMA_UNASSIGNED_HOLDINGS AMRAJ ANKPU APXCP AXJTR BELTK BJAXD BKOJK BLECG BLXMC CS3 DU5 EBS EFJIC EFKBS EFLBG EO8 EO9 EP2 EP3 FDB FIRID FNPLU FYGXN G-Q GBLVA IHE J1W JARJE JJJVA KCYFY KOM LY6 LY7 LY9 M41 MO0 N9A O-L O9- OAUVE OZT P-8 P-9 PC. Q38 RNS ROL RPZ SDF SDG SDP SEN SES SEW SPC SPCBC SSJ SSR SST SSZ T5K ~G- ~HD 9DU AAQXK AAYXX ABWVN ABXDB ACNNM ACRPL ADMUD ADNMO AEGFY AGQPQ AI. ASPBG AVWKF AZFZN CITATION EJD FEDTE FGOYB G-2 HMC HVGLF HZ~ R2- SAC SET VH1 WUQ ZMT |
| ID | FETCH-LOGICAL-c261t-6f0b94557c9ee099d207dc240dec800b53d302d5421c02c7fc7b655a93fb86cf3 |
| ISICitedReferencesCount | 1 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001586723600001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 0360-1323 |
| IngestDate | Sat Nov 29 07:00:56 EST 2025 Sat Oct 25 17:17:18 EDT 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Luminance map Ordinal ranking Personal daylighting preference Convolutional autoencoder (CAE) |
| Language | English |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c261t-6f0b94557c9ee099d207dc240dec800b53d302d5421c02c7fc7b655a93fb86cf3 |
| ORCID | 0000-0002-5181-2944 0000-0001-6788-8372 |
| ParticipantIDs | crossref_primary_10_1016_j_buildenv_2025_113595 elsevier_sciencedirect_doi_10_1016_j_buildenv_2025_113595 |
| PublicationCentury | 2000 |
| PublicationDate | 2025-11-01 2025-11-00 |
| PublicationDateYYYYMMDD | 2025-11-01 |
| PublicationDate_xml | – month: 11 year: 2025 text: 2025-11-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationTitle | Building and environment |
| PublicationYear | 2025 |
| Publisher | Elsevier Ltd |
| Publisher_xml | – name: Elsevier Ltd |
| References | Zou, Zhou, Yang, Spanos (bib0039) 2018; 177 Yoon (bib0046) 2020; 221 Hu, Wang, Tan, Cai (bib0045) 2023; 301 Xiong, Awalgaonkar, Tzempelikos, Bilionis, Performance, Laboratories, Jischke, Lafayette (bib0031) 2020; 181 Fan, Painter, Mardaljevic (bib0013) 2009 Russakovsky, Deng, Su, Krause, Satheesh, Ma, Huang, Karpathy, Khosla, Bernstein (bib0060) 2015; 115 Zaikina, Matusiak, Klöckner (bib0007) 2015; 11 Fan, Xiao, Zhao, Wang (bib0055) 2018; 211 Van Den Wymelenberg, Inanici (bib0005) 2014; 10 Banihashemi, Weber, Lang (bib0035) 2022; 207 Choi, Yoon (bib0043) 2021; 203 Tra, Amayri, Bouguila (bib0048) 2022; 259 Van Den Wymelenberg, Inanici, Johnson (bib0008) 2010; 7 Li, Song, Luo (bib0064) 2017 Moeck (bib0010) 2007; 4 Konis (bib0024) 2014; 77 Kim, Song, Lee, Lee (bib0040) 2024; 311 Gao, Qian, Dong, Liu (bib0037) 2024; 320 Hirning, Isoardi, Garcia-Hansen (bib0022) 2017; 113 Bonfigli, Felicetti, Principi, Fagiani, Squartini, Piazza (bib0056) 2018; 158 Krizhevsky, Sutskever, Hinton (bib0059) 2012; 25 Xiang, Fu, Zhou, Wang, Zhang, Hu, Xu, Liu, Liu, Ma (bib0036) 2021; 187 Kim, Tzempelikos (bib0068) 2022; 213 Xiong, Tzempelikos, Bilionis, Awalgaonkar, Lee, Konstantzos, Sadeghi, Karava (bib0030) 2018; 138 Mahić, Galicinao, Van Den Wymelenberg (bib0025) 2017; 113 Liguori, Quintana, Fu, Miller, Frisch, van Treeck (bib0057) 2024; 310 Li, Li, Yang, Dong (bib0042) 2019; 45 Garretón, Colombo, Pattini (bib0021) 2018; 166 Gautam, Jebelli (bib0053) 2024; 165 de Vries, Heynderickx, de Kort (bib0002) 2022; 54 Suk, Schiler, Kensek (bib0026) 2013; 64 Liang, Kamat, Menassa (bib0054) 2020; 120 (bib0061) 2018 Konstantzos, Tzempelikos (bib0019) 2017; 113 Lindelöf, Morel (bib0029) 2008; 36 Wagdy, Hansen, Isoardi, Allan (bib0020) 2017; 2017 Loy-Benitez, Li, Nam, Yoo (bib0049) 2020; 52 Ware (bib0006) 2012 Inanici (bib0009) 2006; 38 Van Den Wymelenberg, Inanici (bib0032) 2016; 12 Aggarwal, Hinneburg, Keim (bib0034) 2001; 8 Cuttle (bib0003) 2004; 36 Suk, Schiler (bib0015) 2013; 45 Fu, Quintana, Nagy, Miller (bib0050) 2024; 236 Wienold, Christoffersen (bib0017) 2006; 38 Hirning, Isoardi, Cowling (bib0018) 2014; 70 Marty, Fontoynont, Christoffersen (bib0004) 2003 Xiao, Khayatian (bib0038) 2020; 224 Liguori, Markovic, Dam, Frisch, van Treeck, Causone (bib0058) 2021; 191 Wang, Li, Liang, Yoon, Mu, Liu (bib0044) 2023; 290 Kim, Tzempelikos (bib0069) 2021; 129 Xiong, Tzempelikos, Bilionis, Karava (bib0001) 2019; 193 Mah, Tzempelikos (bib0033) 2024; 266 Li, Yuan, Liu, Han, Stouffs (bib0067) 2024; 305 Jiang, Risbeck, Samy, Zhang, Cyrus, Lee (bib0047) 2023; 285 Dai, He, Sun (bib0065) 2015 Liu, Pang, Karlsson, Gong (bib0051) 2020; 183 Araya, Grolinger, ElYamany, Capretz, Bitsuamlak (bib0052) 2017; 144 Kruisselbrink, Dangol, van Loenen (bib0062) 2020; 169 Suk, Schiler, Kensek (bib0027) 2017; 113 Kingsley, C, Brown (bib0028) 2010; 86 Pierson, Cauwerts, Bodart, Wienold (bib0014) 2021; 17 Prechelt (bib0063) 2012 Konis (bib0023) 2013; 59 He, Li, Gao, Chen, Wu, Cheng, Lin (bib0066) 2021; 206 Cauwerts (bib0012) 2018; 4 Yang, Li, Chen, Hong (bib0041) 2022; 84 Wagdy, Garcia-Hansen, Isoardi, Pham (bib0016) 2019; 9 Tyukhova, Waters (bib0011) 2014; 10 Tyukhova (10.1016/j.buildenv.2025.113595_bib0011) 2014; 10 Mah (10.1016/j.buildenv.2025.113595_bib0033) 2024; 266 Bonfigli (10.1016/j.buildenv.2025.113595_bib0056) 2018; 158 Xiong (10.1016/j.buildenv.2025.113595_bib0031) 2020; 181 Garretón (10.1016/j.buildenv.2025.113595_bib0021) 2018; 166 Zaikina (10.1016/j.buildenv.2025.113595_bib0007) 2015; 11 Liu (10.1016/j.buildenv.2025.113595_bib0051) 2020; 183 Liguori (10.1016/j.buildenv.2025.113595_bib0057) 2024; 310 Prechelt (10.1016/j.buildenv.2025.113595_bib0063) 2012 Inanici (10.1016/j.buildenv.2025.113595_bib0009) 2006; 38 Cuttle (10.1016/j.buildenv.2025.113595_bib0003) 2004; 36 Fu (10.1016/j.buildenv.2025.113595_bib0050) 2024; 236 Gao (10.1016/j.buildenv.2025.113595_bib0037) 2024; 320 Wienold (10.1016/j.buildenv.2025.113595_bib0017) 2006; 38 Xiong (10.1016/j.buildenv.2025.113595_bib0001) 2019; 193 Cauwerts (10.1016/j.buildenv.2025.113595_bib0012) 2018; 4 Fan (10.1016/j.buildenv.2025.113595_bib0055) 2018; 211 Moeck (10.1016/j.buildenv.2025.113595_bib0010) 2007; 4 Tra (10.1016/j.buildenv.2025.113595_bib0048) 2022; 259 Suk (10.1016/j.buildenv.2025.113595_bib0015) 2013; 45 Banihashemi (10.1016/j.buildenv.2025.113595_bib0035) 2022; 207 Ware (10.1016/j.buildenv.2025.113595_bib0006) 2012 Araya (10.1016/j.buildenv.2025.113595_bib0052) 2017; 144 Yang (10.1016/j.buildenv.2025.113595_bib0041) 2022; 84 Kim (10.1016/j.buildenv.2025.113595_bib0040) 2024; 311 Suk (10.1016/j.buildenv.2025.113595_bib0027) 2017; 113 Li (10.1016/j.buildenv.2025.113595_bib0067) 2024; 305 Marty (10.1016/j.buildenv.2025.113595_bib0004) 2003 Van Den Wymelenberg (10.1016/j.buildenv.2025.113595_bib0008) 2010; 7 Choi (10.1016/j.buildenv.2025.113595_bib0043) 2021; 203 Gautam (10.1016/j.buildenv.2025.113595_bib0053) 2024; 165 Liguori (10.1016/j.buildenv.2025.113595_bib0058) 2021; 191 He (10.1016/j.buildenv.2025.113595_bib0066) 2021; 206 Jiang (10.1016/j.buildenv.2025.113595_bib0047) 2023; 285 Krizhevsky (10.1016/j.buildenv.2025.113595_bib0059) 2012; 25 Liang (10.1016/j.buildenv.2025.113595_bib0054) 2020; 120 Zou (10.1016/j.buildenv.2025.113595_bib0039) 2018; 177 Konis (10.1016/j.buildenv.2025.113595_bib0024) 2014; 77 Xiong (10.1016/j.buildenv.2025.113595_bib0030) 2018; 138 Suk (10.1016/j.buildenv.2025.113595_bib0026) 2013; 64 Kruisselbrink (10.1016/j.buildenv.2025.113595_bib0062) 2020; 169 de Vries (10.1016/j.buildenv.2025.113595_bib0002) 2022; 54 Loy-Benitez (10.1016/j.buildenv.2025.113595_bib0049) 2020; 52 Pierson (10.1016/j.buildenv.2025.113595_bib0014) 2021; 17 Mahić (10.1016/j.buildenv.2025.113595_bib0025) 2017; 113 Wagdy (10.1016/j.buildenv.2025.113595_bib0016) 2019; 9 Kim (10.1016/j.buildenv.2025.113595_bib0068) 2022; 213 Van Den Wymelenberg (10.1016/j.buildenv.2025.113595_bib0032) 2016; 12 Li (10.1016/j.buildenv.2025.113595_bib0042) 2019; 45 Li (10.1016/j.buildenv.2025.113595_bib0064) 2017 Lindelöf (10.1016/j.buildenv.2025.113595_bib0029) 2008; 36 Kingsley (10.1016/j.buildenv.2025.113595_bib0028) 2010; 86 Konstantzos (10.1016/j.buildenv.2025.113595_bib0019) 2017; 113 Xiao (10.1016/j.buildenv.2025.113595_bib0038) 2020; 224 Yoon (10.1016/j.buildenv.2025.113595_bib0046) 2020; 221 Fan (10.1016/j.buildenv.2025.113595_bib0013) 2009 Hirning (10.1016/j.buildenv.2025.113595_bib0018) 2014; 70 Wagdy (10.1016/j.buildenv.2025.113595_bib0020) 2017; 2017 Aggarwal (10.1016/j.buildenv.2025.113595_bib0034) 2001; 8 Xiang (10.1016/j.buildenv.2025.113595_bib0036) 2021; 187 Hirning (10.1016/j.buildenv.2025.113595_bib0022) 2017; 113 Konis (10.1016/j.buildenv.2025.113595_bib0023) 2013; 59 Van Den Wymelenberg (10.1016/j.buildenv.2025.113595_bib0005) 2014; 10 Wang (10.1016/j.buildenv.2025.113595_bib0044) 2023; 290 Hu (10.1016/j.buildenv.2025.113595_bib0045) 2023; 301 Russakovsky (10.1016/j.buildenv.2025.113595_bib0060) 2015; 115 Kim (10.1016/j.buildenv.2025.113595_bib0069) 2021; 129 Dai (10.1016/j.buildenv.2025.113595_bib0065) 2015 |
| References_xml | – volume: 211 start-page: 1123 year: 2018 end-page: 1135 ident: bib0055 article-title: Analytical investigation of autoencoder-based methods for unsupervised anomaly detection in building energy data publication-title: Appl. Energy. – volume: 36 start-page: 83 year: 2008 end-page: 96 ident: bib0029 article-title: Bayesian estimation of visual discomfort publication-title: Build. Res. Inf. – volume: 11 start-page: 193 year: 2015 end-page: 207 ident: bib0007 article-title: Luminance-based measures of shape and detail distinctness of 3D objects as important predictors of light modeling concept. Results of a full-scale study pairing proposed measures with subjective responses publication-title: LEUKOS - J. Illum. Eng. Soc. North Am. – start-page: 3617 year: 2017 end-page: 3625 ident: bib0064 article-title: Improving pairwise ranking for multi-label image classification publication-title: Proc. IEEE Conf. Comput. Vis. Pattern Recognit – volume: 4 year: 2018 ident: bib0012 article-title: Application of high-dynamic range imaging techniques in architecture : a step toward high-quality daylit interiors? publication-title: J. Imaging. – volume: 166 start-page: 145 year: 2018 end-page: 153 ident: bib0021 article-title: A global evaluation of discomfort glare metrics in real office spaces with presence of direct sunlight publication-title: Energy Build. – volume: 305 year: 2024 ident: bib0067 article-title: A predictive model for daylight performance based on multimodal generative adversarial networks at the early design stage publication-title: Energy Build. – volume: 9 start-page: 1 year: 2019 end-page: 24 ident: bib0016 article-title: A parametric method for remapping and calibrating fisheye images for glare analysis publication-title: Buildings – volume: 193 start-page: 111 year: 2019 end-page: 126 ident: bib0001 article-title: A personalized daylighting control approach to dynamically optimize visual satisfaction and lighting energy use publication-title: Energy Build. – volume: 45 start-page: 450 year: 2013 end-page: 463 ident: bib0015 article-title: Investigation of Evalglare software, daylight glare probability and high dynamic range imaging for daylight glare analysis publication-title: Light. Res. Technol. – volume: 77 start-page: 67 year: 2014 end-page: 79 ident: bib0024 article-title: Predicting visual comfort in side-lit open-plan core zones: results of a field study pairing high dynamic range images with subjective responses publication-title: Energy Build. – volume: 177 start-page: 12 year: 2018 end-page: 22 ident: bib0039 article-title: Towards occupant activity driven smart buildings via WiFi-enabled IoT devices and deep learning publication-title: Energy Build. – start-page: 53 year: 2012 end-page: 67 ident: bib0063 article-title: Early stopping — But when? publication-title: BT - Neural Networks: Tricks of the Trade – volume: 285 year: 2023 ident: bib0047 article-title: A timeseries supervised learning framework for fault prediction in chiller systems publication-title: Energy Build. – volume: 183 year: 2020 ident: bib0051 article-title: Anomaly detection based on machine learning in IoT-based vertical plant wall for indoor climate control publication-title: Build. Environ. – volume: 138 start-page: 74 year: 2018 end-page: 88 ident: bib0030 article-title: Inferring personalized visual satisfaction profiles in daylit offices from comparative preferences using a Bayesian approach publication-title: Build. Environ. – volume: 206 year: 2021 ident: bib0066 article-title: Predictive models for daylight performance of general floorplans based on CNN and GAN: a proof-of-concept study publication-title: Build. Environ. – volume: 10 start-page: 87 year: 2014 end-page: 99 ident: bib0011 article-title: An assessment of high dynamic range luminance measurements with LED lighting publication-title: LEUKOS - J. Illum. Eng. Soc. North Am. – volume: 158 start-page: 1461 year: 2018 end-page: 1474 ident: bib0056 article-title: Denoising autoencoders for non-intrusive load monitoring: improvements and comparative evaluation publication-title: Energy Build. – volume: 86 start-page: 530 year: 2010 end-page: 544 ident: bib0028 article-title: Preference uncertainty, preference learning, and paired comparison experiments publication-title: Land. Econ. – volume: 259 year: 2022 ident: bib0048 article-title: Outlier detection via multiclass deep autoencoding gaussian mixture model for building chiller diagnosis publication-title: Energy Build. – volume: 115 start-page: 211 year: 2015 end-page: 252 ident: bib0060 article-title: Imagenet large scale visual recognition challenge publication-title: Int. J. Comput. Vis. – volume: 266 year: 2024 ident: bib0033 article-title: Inferring personal daylighting preferences using HDRI and deep learning techniques publication-title: Build. Environ. – volume: 310 year: 2024 ident: bib0057 article-title: Opening the Black Box: towards inherently interpretable energy data imputation models using building physics insight publication-title: Energy Build. – volume: 59 start-page: 662 year: 2013 end-page: 677 ident: bib0023 article-title: Evaluating daylighting effectiveness and occupant visual comfort in a side-lit open-plan office building in San Francisco, California publication-title: Build. Environ. – volume: 25 year: 2012 ident: bib0059 article-title: Imagenet classification with deep convolutional neural networks publication-title: Adv. Neural Inf. Process. Syst. – volume: 38 start-page: 743 year: 2006 end-page: 757 ident: bib0017 article-title: Evaluation methods and development of a new glare prediction model for daylight environments with the use of CCD cameras publication-title: Energy Build. – volume: 12 start-page: 113 year: 2016 end-page: 138 ident: bib0032 article-title: Evaluating a new suite of luminance-based design metrics for predicting human visual comfort in offices with daylight publication-title: Leukos. – volume: 129 year: 2021 ident: bib0069 article-title: Semi-automated luminance map re-projection via high dynamic range imaging and indoor space 3-D reconstruction publication-title: Autom. Constr. – volume: 207 year: 2022 ident: bib0035 article-title: Model order reduction of building energy simulation models using a convolutional neural network autoencoder publication-title: Build. Environ. – volume: 38 start-page: 135 year: 2006 ident: bib0009 article-title: Evaluation of high dynamic range photography as a luminance data acquisition system publication-title: Light. Res. Technol. – volume: 311 year: 2024 ident: bib0040 article-title: Time-series data clustering with load-shape preservation for identifying residential energy consumption behaviors publication-title: Energy Build. – volume: 165 year: 2024 ident: bib0053 article-title: Autoencoder-based photoplethysmography (PPG) signal reliability enhancement in construction health monitoring publication-title: Autom. Constr. – volume: 84 year: 2022 ident: bib0041 article-title: Characterizing residential load patterns on multi-time scales utilizing LSTM autoencoder and electricity consumption data publication-title: Sustain. Cities Soc. – start-page: 69 year: 2012 end-page: 95 ident: bib0006 article-title: Lightness, Brightness, Contrast, and Constancy, Inf. Vis. Percept. Des – volume: 169 year: 2020 ident: bib0062 article-title: Recommendations for long-term luminance distribution measurements: the spatial resolution publication-title: Build. Environ. – volume: 17 start-page: 140 year: 2021 end-page: 169 ident: bib0014 article-title: Tutorial: luminance maps for daylighting studies from high dynamic range photography publication-title: Leukos. – volume: 52 year: 2020 ident: bib0049 article-title: Sustainable subway indoor air quality monitoring and fault-tolerant ventilation control using a sparse autoencoder-driven sensor self-validation, Sustain publication-title: Cities Soc. – start-page: 251 year: 2009 end-page: 256 ident: bib0013 article-title: A data collection method for long-term field studies of visual comfort in real-world daylit office environments publication-title: Proc. PLEA – volume: 203 year: 2021 ident: bib0043 article-title: Autoencoder-driven fault detection and diagnosis in building automation systems: residual-based and latent space-based approaches publication-title: Build. Environ. – year: 2018 ident: bib0061 article-title: BS EN 17037:2018: daylight in buildings – volume: 4 start-page: 99 year: 2007 end-page: 112 ident: bib0010 article-title: Accuracy of luminance maps obtained from high dynamic range images publication-title: Leukos. – volume: 64 start-page: 113 year: 2013 end-page: 122 ident: bib0026 article-title: Development of new daylight glare analysis methodology using absolute glare factor and relative glare factor publication-title: Energy Build. – volume: 10 start-page: 145 year: 2014 end-page: 164 ident: bib0005 article-title: A critical investigation of common lighting design metrics for predicting human visual comfort in offices with daylight publication-title: Leukos. – volume: 54 start-page: 798 year: 2022 end-page: 818 ident: bib0002 article-title: From luminance to brightness: a data-driven approach to support brightness assessments in open plan offices publication-title: Light. Res. Technol. – volume: 7 start-page: 103 year: 2010 end-page: 122 ident: bib0008 article-title: The effect of luminance distribution patterns on occupant preference in a daylit office environment publication-title: Leukos. – volume: 45 start-page: 588 year: 2019 end-page: 595 ident: bib0042 article-title: Indoor tracking trajectory data similarity analysis with a deep convolutional autoencoder publication-title: Sustain. Cities Soc. – volume: 2017 start-page: 3419 year: 2017 end-page: 3426 ident: bib0020 article-title: Multi-region contrast method - A new framework for post-processing HDRI luminance information for visual discomfort analysis publication-title: Proc. 33rd PLEA Int. Conf. Des. to Thrive – volume: 113 start-page: 121 year: 2017 end-page: 130 ident: bib0027 article-title: Investigation of existing discomfort glare indices using human subject study data publication-title: Build. Environ. – volume: 181 year: 2020 ident: bib0031 article-title: Efficient learning of personalized visual preferences in daylit offices : an online elicitation framework publication-title: Build. Environ. – volume: 236 year: 2024 ident: bib0050 article-title: Filling time-series gaps using image techniques: multidimensional context autoencoder approach for building energy data imputation publication-title: Appl. Therm. Eng. – volume: 36 start-page: 201 year: 2004 end-page: 216 ident: bib0003 article-title: Brightness, lightness, and providing “a preconceived appearance to the interior publication-title: Light. Res. Technol. – volume: 113 start-page: 65 year: 2017 end-page: 77 ident: bib0019 article-title: Daylight glare evaluation with the sun in the field of view through window shades publication-title: Build. Environ. – volume: 113 start-page: 78 year: 2017 end-page: 91 ident: bib0025 article-title: A pilot daylighting field study: testing the usefulness of laboratory-derived luminance-based metrics for building design and control publication-title: Build. Environ. – volume: 320 year: 2024 ident: bib0037 article-title: Machine learning-based reduced-order reconstruction method for flow fields publication-title: Energy Build. – volume: 113 start-page: 107 year: 2017 end-page: 120 ident: bib0022 article-title: Prediction of discomfort glare from windows under tropical skies publication-title: Build. Environ. – volume: 8 start-page: 420 year: 2001 end-page: 434 ident: bib0034 article-title: On the surprising behavior of distance metrics in high dimensional space publication-title: Database Theory—ICDT 2001 8th Int. Conf – volume: 187 year: 2021 ident: bib0036 article-title: Non-intrusive reduced order model of urban airflow with dynamic boundary conditions publication-title: Build. Environ – volume: 290 year: 2023 ident: bib0044 article-title: Fault detection and calibration for building energy system using bayesian inference and sparse autoencoder: a case study in photovoltaic thermal heat pump system publication-title: Energy Build. – volume: 221 year: 2020 ident: bib0046 article-title: In-situ sensor calibration in an operational air-handling unit coupling autoencoder and bayesian inference publication-title: Energy Build. – volume: 70 start-page: 427 year: 2014 end-page: 440 ident: bib0018 article-title: Discomfort glare in open plan green buildings publication-title: Energy Build. – volume: 213 year: 2022 ident: bib0068 article-title: Performance evaluation of non-intrusive luminance mapping towards human-centered daylighting control publication-title: Build. Environ. – volume: 120 year: 2020 ident: bib0054 article-title: Teaching robots to perform quasi-repetitive construction tasks through human demonstration publication-title: Autom. Constr. – volume: 301 year: 2023 ident: bib0045 article-title: Digital twin-enhanced predictive maintenance for indoor climate: a parallel LSTM-autoencoder failure prediction approach publication-title: Energy Build. – volume: 224 year: 2020 ident: bib0038 article-title: Unsupervised learning for feature projection: extracting patterns from multidimensional building measurements publication-title: Energy Build. – volume: 191 year: 2021 ident: bib0058 article-title: Indoor environment data time-series reconstruction using autoencoder neural networks publication-title: Build. Environ. – start-page: 3992 year: 2015 end-page: 4000 ident: bib0065 article-title: Convolutional feature masking for joint object and stuff segmentation publication-title: Proc. IEEE Conf. Comput. Vis. Pattern Recognit – year: 2003 ident: bib0004 article-title: User assessment of visual comfort : review of existing methods publication-title: ECCO-Build. – volume: 144 start-page: 191 year: 2017 end-page: 206 ident: bib0052 article-title: An ensemble learning framework for anomaly detection in building energy consumption publication-title: Energy Build. – volume: 259 year: 2022 ident: 10.1016/j.buildenv.2025.113595_bib0048 article-title: Outlier detection via multiclass deep autoencoding gaussian mixture model for building chiller diagnosis publication-title: Energy Build. doi: 10.1016/j.enbuild.2022.111893 – volume: 129 year: 2021 ident: 10.1016/j.buildenv.2025.113595_bib0069 article-title: Semi-automated luminance map re-projection via high dynamic range imaging and indoor space 3-D reconstruction publication-title: Autom. Constr. doi: 10.1016/j.autcon.2021.103812 – start-page: 3617 year: 2017 ident: 10.1016/j.buildenv.2025.113595_bib0064 article-title: Improving pairwise ranking for multi-label image classification – volume: 224 year: 2020 ident: 10.1016/j.buildenv.2025.113595_bib0038 article-title: Unsupervised learning for feature projection: extracting patterns from multidimensional building measurements publication-title: Energy Build. doi: 10.1016/j.enbuild.2020.110228 – volume: 183 year: 2020 ident: 10.1016/j.buildenv.2025.113595_bib0051 article-title: Anomaly detection based on machine learning in IoT-based vertical plant wall for indoor climate control publication-title: Build. Environ. doi: 10.1016/j.buildenv.2020.107212 – volume: 77 start-page: 67 year: 2014 ident: 10.1016/j.buildenv.2025.113595_bib0024 article-title: Predicting visual comfort in side-lit open-plan core zones: results of a field study pairing high dynamic range images with subjective responses publication-title: Energy Build. doi: 10.1016/j.enbuild.2014.03.035 – volume: 221 year: 2020 ident: 10.1016/j.buildenv.2025.113595_bib0046 article-title: In-situ sensor calibration in an operational air-handling unit coupling autoencoder and bayesian inference publication-title: Energy Build. doi: 10.1016/j.enbuild.2020.110026 – volume: 203 year: 2021 ident: 10.1016/j.buildenv.2025.113595_bib0043 article-title: Autoencoder-driven fault detection and diagnosis in building automation systems: residual-based and latent space-based approaches publication-title: Build. Environ. doi: 10.1016/j.buildenv.2021.108066 – volume: 54 start-page: 798 year: 2022 ident: 10.1016/j.buildenv.2025.113595_bib0002 article-title: From luminance to brightness: a data-driven approach to support brightness assessments in open plan offices publication-title: Light. Res. Technol. doi: 10.1177/14771535221117365 – volume: 193 start-page: 111 year: 2019 ident: 10.1016/j.buildenv.2025.113595_bib0001 article-title: A personalized daylighting control approach to dynamically optimize visual satisfaction and lighting energy use publication-title: Energy Build. doi: 10.1016/j.enbuild.2019.03.046 – volume: 113 start-page: 65 year: 2017 ident: 10.1016/j.buildenv.2025.113595_bib0019 article-title: Daylight glare evaluation with the sun in the field of view through window shades publication-title: Build. Environ. doi: 10.1016/j.buildenv.2016.09.009 – volume: 36 start-page: 83 year: 2008 ident: 10.1016/j.buildenv.2025.113595_bib0029 article-title: Bayesian estimation of visual discomfort publication-title: Build. Res. Inf. doi: 10.1080/09613210701544061 – volume: 113 start-page: 107 year: 2017 ident: 10.1016/j.buildenv.2025.113595_bib0022 article-title: Prediction of discomfort glare from windows under tropical skies publication-title: Build. Environ. doi: 10.1016/j.buildenv.2016.08.005 – volume: 59 start-page: 662 year: 2013 ident: 10.1016/j.buildenv.2025.113595_bib0023 article-title: Evaluating daylighting effectiveness and occupant visual comfort in a side-lit open-plan office building in San Francisco, California publication-title: Build. Environ. doi: 10.1016/j.buildenv.2012.09.017 – volume: 64 start-page: 113 year: 2013 ident: 10.1016/j.buildenv.2025.113595_bib0026 article-title: Development of new daylight glare analysis methodology using absolute glare factor and relative glare factor publication-title: Energy Build. doi: 10.1016/j.enbuild.2013.04.020 – volume: 311 year: 2024 ident: 10.1016/j.buildenv.2025.113595_bib0040 article-title: Time-series data clustering with load-shape preservation for identifying residential energy consumption behaviors publication-title: Energy Build. doi: 10.1016/j.enbuild.2024.114130 – volume: 301 year: 2023 ident: 10.1016/j.buildenv.2025.113595_bib0045 article-title: Digital twin-enhanced predictive maintenance for indoor climate: a parallel LSTM-autoencoder failure prediction approach publication-title: Energy Build. doi: 10.1016/j.enbuild.2023.113738 – volume: 285 year: 2023 ident: 10.1016/j.buildenv.2025.113595_bib0047 article-title: A timeseries supervised learning framework for fault prediction in chiller systems publication-title: Energy Build. doi: 10.1016/j.enbuild.2023.112876 – volume: 2017 start-page: 3419 year: 2017 ident: 10.1016/j.buildenv.2025.113595_bib0020 article-title: Multi-region contrast method - A new framework for post-processing HDRI luminance information for visual discomfort analysis – volume: 305 year: 2024 ident: 10.1016/j.buildenv.2025.113595_bib0067 article-title: A predictive model for daylight performance based on multimodal generative adversarial networks at the early design stage publication-title: Energy Build. doi: 10.1016/j.enbuild.2023.113876 – volume: 36 start-page: 201 year: 2004 ident: 10.1016/j.buildenv.2025.113595_bib0003 article-title: Brightness, lightness, and providing “a preconceived appearance to the interior publication-title: Light. Res. Technol. doi: 10.1191/1365782804li115oa – volume: 25 year: 2012 ident: 10.1016/j.buildenv.2025.113595_bib0059 article-title: Imagenet classification with deep convolutional neural networks publication-title: Adv. Neural Inf. Process. Syst. – volume: 166 start-page: 145 year: 2018 ident: 10.1016/j.buildenv.2025.113595_bib0021 article-title: A global evaluation of discomfort glare metrics in real office spaces with presence of direct sunlight publication-title: Energy Build. doi: 10.1016/j.enbuild.2018.01.024 – volume: 10 start-page: 87 year: 2014 ident: 10.1016/j.buildenv.2025.113595_bib0011 article-title: An assessment of high dynamic range luminance measurements with LED lighting publication-title: LEUKOS - J. Illum. Eng. Soc. North Am. – volume: 38 start-page: 135 year: 2006 ident: 10.1016/j.buildenv.2025.113595_bib0009 article-title: Evaluation of high dynamic range photography as a luminance data acquisition system publication-title: Light. Res. Technol. doi: 10.1177/136578280603800210 – volume: 70 start-page: 427 year: 2014 ident: 10.1016/j.buildenv.2025.113595_bib0018 article-title: Discomfort glare in open plan green buildings publication-title: Energy Build. doi: 10.1016/j.enbuild.2013.11.053 – volume: 165 year: 2024 ident: 10.1016/j.buildenv.2025.113595_bib0053 article-title: Autoencoder-based photoplethysmography (PPG) signal reliability enhancement in construction health monitoring publication-title: Autom. Constr. doi: 10.1016/j.autcon.2024.105537 – start-page: 69 year: 2012 ident: 10.1016/j.buildenv.2025.113595_bib0006 – volume: 191 year: 2021 ident: 10.1016/j.buildenv.2025.113595_bib0058 article-title: Indoor environment data time-series reconstruction using autoencoder neural networks publication-title: Build. Environ. doi: 10.1016/j.buildenv.2021.107623 – volume: 181 year: 2020 ident: 10.1016/j.buildenv.2025.113595_bib0031 article-title: Efficient learning of personalized visual preferences in daylit offices : an online elicitation framework publication-title: Build. Environ. doi: 10.1016/j.buildenv.2020.107013 – volume: 4 year: 2018 ident: 10.1016/j.buildenv.2025.113595_bib0012 article-title: Application of high-dynamic range imaging techniques in architecture : a step toward high-quality daylit interiors? publication-title: J. Imaging. doi: 10.3390/jimaging4010019 – start-page: 251 year: 2009 ident: 10.1016/j.buildenv.2025.113595_bib0013 article-title: A data collection method for long-term field studies of visual comfort in real-world daylit office environments – volume: 8 start-page: 420 year: 2001 ident: 10.1016/j.buildenv.2025.113595_bib0034 article-title: On the surprising behavior of distance metrics in high dimensional space – volume: 113 start-page: 121 year: 2017 ident: 10.1016/j.buildenv.2025.113595_bib0027 article-title: Investigation of existing discomfort glare indices using human subject study data publication-title: Build. Environ. doi: 10.1016/j.buildenv.2016.09.018 – volume: 236 year: 2024 ident: 10.1016/j.buildenv.2025.113595_bib0050 article-title: Filling time-series gaps using image techniques: multidimensional context autoencoder approach for building energy data imputation publication-title: Appl. Therm. Eng. doi: 10.1016/j.applthermaleng.2023.121545 – volume: 213 year: 2022 ident: 10.1016/j.buildenv.2025.113595_bib0068 article-title: Performance evaluation of non-intrusive luminance mapping towards human-centered daylighting control publication-title: Build. Environ. doi: 10.1016/j.buildenv.2022.108857 – volume: 169 year: 2020 ident: 10.1016/j.buildenv.2025.113595_bib0062 article-title: Recommendations for long-term luminance distribution measurements: the spatial resolution publication-title: Build. Environ. doi: 10.1016/j.buildenv.2019.106538 – volume: 120 year: 2020 ident: 10.1016/j.buildenv.2025.113595_bib0054 article-title: Teaching robots to perform quasi-repetitive construction tasks through human demonstration publication-title: Autom. Constr. doi: 10.1016/j.autcon.2020.103370 – volume: 144 start-page: 191 year: 2017 ident: 10.1016/j.buildenv.2025.113595_bib0052 article-title: An ensemble learning framework for anomaly detection in building energy consumption publication-title: Energy Build. doi: 10.1016/j.enbuild.2017.02.058 – volume: 38 start-page: 743 year: 2006 ident: 10.1016/j.buildenv.2025.113595_bib0017 article-title: Evaluation methods and development of a new glare prediction model for daylight environments with the use of CCD cameras publication-title: Energy Build. doi: 10.1016/j.enbuild.2006.03.017 – start-page: 53 year: 2012 ident: 10.1016/j.buildenv.2025.113595_bib0063 article-title: Early stopping — But when? – volume: 10 start-page: 145 year: 2014 ident: 10.1016/j.buildenv.2025.113595_bib0005 article-title: A critical investigation of common lighting design metrics for predicting human visual comfort in offices with daylight publication-title: Leukos. doi: 10.1080/15502724.2014.881720 – volume: 207 year: 2022 ident: 10.1016/j.buildenv.2025.113595_bib0035 article-title: Model order reduction of building energy simulation models using a convolutional neural network autoencoder publication-title: Build. Environ. doi: 10.1016/j.buildenv.2021.108498 – volume: 290 year: 2023 ident: 10.1016/j.buildenv.2025.113595_bib0044 article-title: Fault detection and calibration for building energy system using bayesian inference and sparse autoencoder: a case study in photovoltaic thermal heat pump system publication-title: Energy Build. doi: 10.1016/j.enbuild.2023.113051 – volume: 187 year: 2021 ident: 10.1016/j.buildenv.2025.113595_bib0036 article-title: Non-intrusive reduced order model of urban airflow with dynamic boundary conditions publication-title: Build. Environ doi: 10.1016/j.buildenv.2020.107397 – volume: 211 start-page: 1123 year: 2018 ident: 10.1016/j.buildenv.2025.113595_bib0055 article-title: Analytical investigation of autoencoder-based methods for unsupervised anomaly detection in building energy data publication-title: Appl. Energy. doi: 10.1016/j.apenergy.2017.12.005 – volume: 12 start-page: 113 year: 2016 ident: 10.1016/j.buildenv.2025.113595_bib0032 article-title: Evaluating a new suite of luminance-based design metrics for predicting human visual comfort in offices with daylight publication-title: Leukos. doi: 10.1080/15502724.2015.1062392 – volume: 310 year: 2024 ident: 10.1016/j.buildenv.2025.113595_bib0057 article-title: Opening the Black Box: towards inherently interpretable energy data imputation models using building physics insight publication-title: Energy Build. doi: 10.1016/j.enbuild.2024.114071 – volume: 45 start-page: 588 year: 2019 ident: 10.1016/j.buildenv.2025.113595_bib0042 article-title: Indoor tracking trajectory data similarity analysis with a deep convolutional autoencoder publication-title: Sustain. Cities Soc. doi: 10.1016/j.scs.2018.12.025 – volume: 7 start-page: 103 year: 2010 ident: 10.1016/j.buildenv.2025.113595_bib0008 article-title: The effect of luminance distribution patterns on occupant preference in a daylit office environment publication-title: Leukos. doi: 10.1582/LEUKOS.2010.07.02003 – volume: 9 start-page: 1 year: 2019 ident: 10.1016/j.buildenv.2025.113595_bib0016 article-title: A parametric method for remapping and calibrating fisheye images for glare analysis publication-title: Buildings doi: 10.3390/buildings9100219 – volume: 113 start-page: 78 year: 2017 ident: 10.1016/j.buildenv.2025.113595_bib0025 article-title: A pilot daylighting field study: testing the usefulness of laboratory-derived luminance-based metrics for building design and control publication-title: Build. Environ. doi: 10.1016/j.buildenv.2016.11.024 – volume: 84 year: 2022 ident: 10.1016/j.buildenv.2025.113595_bib0041 article-title: Characterizing residential load patterns on multi-time scales utilizing LSTM autoencoder and electricity consumption data publication-title: Sustain. Cities Soc. doi: 10.1016/j.scs.2022.104007 – volume: 11 start-page: 193 year: 2015 ident: 10.1016/j.buildenv.2025.113595_bib0007 article-title: Luminance-based measures of shape and detail distinctness of 3D objects as important predictors of light modeling concept. Results of a full-scale study pairing proposed measures with subjective responses publication-title: LEUKOS - J. Illum. Eng. Soc. North Am. – volume: 266 year: 2024 ident: 10.1016/j.buildenv.2025.113595_bib0033 article-title: Inferring personal daylighting preferences using HDRI and deep learning techniques publication-title: Build. Environ. doi: 10.1016/j.buildenv.2024.112128 – volume: 115 start-page: 211 year: 2015 ident: 10.1016/j.buildenv.2025.113595_bib0060 article-title: Imagenet large scale visual recognition challenge publication-title: Int. J. Comput. Vis. doi: 10.1007/s11263-015-0816-y – start-page: 3992 year: 2015 ident: 10.1016/j.buildenv.2025.113595_bib0065 article-title: Convolutional feature masking for joint object and stuff segmentation – volume: 52 year: 2020 ident: 10.1016/j.buildenv.2025.113595_bib0049 article-title: Sustainable subway indoor air quality monitoring and fault-tolerant ventilation control using a sparse autoencoder-driven sensor self-validation, Sustain publication-title: Cities Soc. – volume: 158 start-page: 1461 year: 2018 ident: 10.1016/j.buildenv.2025.113595_bib0056 article-title: Denoising autoencoders for non-intrusive load monitoring: improvements and comparative evaluation publication-title: Energy Build. doi: 10.1016/j.enbuild.2017.11.054 – volume: 17 start-page: 140 year: 2021 ident: 10.1016/j.buildenv.2025.113595_bib0014 article-title: Tutorial: luminance maps for daylighting studies from high dynamic range photography publication-title: Leukos. doi: 10.1080/15502724.2019.1684319 – volume: 320 year: 2024 ident: 10.1016/j.buildenv.2025.113595_bib0037 article-title: Machine learning-based reduced-order reconstruction method for flow fields publication-title: Energy Build. doi: 10.1016/j.enbuild.2024.114575 – volume: 206 year: 2021 ident: 10.1016/j.buildenv.2025.113595_bib0066 article-title: Predictive models for daylight performance of general floorplans based on CNN and GAN: a proof-of-concept study publication-title: Build. Environ. doi: 10.1016/j.buildenv.2021.108346 – volume: 138 start-page: 74 year: 2018 ident: 10.1016/j.buildenv.2025.113595_bib0030 article-title: Inferring personalized visual satisfaction profiles in daylit offices from comparative preferences using a Bayesian approach publication-title: Build. Environ. doi: 10.1016/j.buildenv.2018.04.022 – volume: 45 start-page: 450 year: 2013 ident: 10.1016/j.buildenv.2025.113595_bib0015 article-title: Investigation of Evalglare software, daylight glare probability and high dynamic range imaging for daylight glare analysis publication-title: Light. Res. Technol. doi: 10.1177/1477153512458671 – volume: 4 start-page: 99 year: 2007 ident: 10.1016/j.buildenv.2025.113595_bib0010 article-title: Accuracy of luminance maps obtained from high dynamic range images publication-title: Leukos. doi: 10.1582/LEUKOS.2007.04.02.002 – volume: 177 start-page: 12 year: 2018 ident: 10.1016/j.buildenv.2025.113595_bib0039 article-title: Towards occupant activity driven smart buildings via WiFi-enabled IoT devices and deep learning publication-title: Energy Build. doi: 10.1016/j.enbuild.2018.08.010 – year: 2003 ident: 10.1016/j.buildenv.2025.113595_bib0004 article-title: User assessment of visual comfort : review of existing methods publication-title: ECCO-Build. – volume: 86 start-page: 530 year: 2010 ident: 10.1016/j.buildenv.2025.113595_bib0028 article-title: Preference uncertainty, preference learning, and paired comparison experiments publication-title: Land. Econ. doi: 10.3368/le.86.3.530 |
| SSID | ssj0016934 |
| Score | 2.4671872 |
| Snippet | •Novel method for leaning personal binary daylighting preference.•Two-stage training method: feature extraction module and preference inference... |
| SourceID | crossref elsevier |
| SourceType | Index Database Publisher |
| StartPage | 113595 |
| SubjectTerms | Convolutional autoencoder (CAE) Luminance map Ordinal ranking Personal daylighting preference |
| Title | A convolutional autoencoder-based method for learning and ranking personal daylighting preference |
| URI | https://dx.doi.org/10.1016/j.buildenv.2025.113595 |
| Volume | 285 |
| WOSCitedRecordID | wos001586723600001&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 issn: 0360-1323 databaseCode: AIEXJ dateStart: 19950101 customDbUrl: isFulltext: true dateEnd: 99991231 titleUrlDefault: https://www.sciencedirect.com omitProxy: false ssIdentifier: ssj0016934 providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1La9wwEBbbTQ_toaQvsm0SdOiteGtLlh9HE1KaHEKhKezNWBq53e3iDfsIob--I0tae5uFtoRehDHowXxCmhnNfEPIuzCBREMtA42XVRADywLJOUNAahmLSEUx1G2xifTqKptM8s-DwTaL_3aeNk12d5ff_Feo8R-CbVJn_wHu7aD4A78RdGwRdmz_CviijSR3UxgmgM16YdgqQS8Dc2eBqxrdBhjOvWekDTOv2joKhsq4VdDfA5rzxni3GeueknbnHdhV1W7795LmOkf3d6umN99mmy7m96dGbX0-_WFj_Arjvq9W08WOD4IJl4y3dYz55JguEskmZIUBGru8f9gyW6Dn3sFtfQizsTTLxuWOzTSm4IywNTh_I8X-YgY3Y6MGZ0jw4kfkgKUiz4bkoLg4n1xuX5KSnDsKMbuYXpb4_tn2Kyg9peP6kDxz1gItLMrPyUA3L8jTHofkS1IVdAdveg9vavGmiDf1eFPEizq8qceb9vCmHd6vyNeP59dnnwJXNyNQaA-vg6QOZR4Lkapca7QAgIUpKFTdQCu0D6TgwEMGImaRCplKa5XKRIgq57XMElXz12TYLBp9RGjIVSQTbl7rqzgCyCBGixQgjVhVJZUekQ9eWOWNpUcpfdzgrPTiLY14SyveEcm9TEun5FnlrcSt8Ie-bx7Q9y150u3cYzJcLzf6hDxWt-vpannqds0vOmmAAw |
| 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=A+convolutional+autoencoder-based+method+for+learning+and+ranking+personal+daylighting+preference&rft.jtitle=Building+and+environment&rft.au=Mah%2C+Dongjun&rft.au=Tzempelikos%2C+Athanasios&rft.date=2025-11-01&rft.pub=Elsevier+Ltd&rft.issn=0360-1323&rft.volume=285&rft_id=info:doi/10.1016%2Fj.buildenv.2025.113595&rft.externalDocID=S0360132325010674 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0360-1323&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0360-1323&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0360-1323&client=summon |