{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T11:00:07Z","timestamp":1781175607420,"version":"3.54.1"},"reference-count":51,"publisher":"Wiley","license":[{"start":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T00:00:00Z","timestamp":1781136000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"},{"start":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T00:00:00Z","timestamp":1781136000000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Computer Graphics Forum"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>As Deep Learning models become increasingly complex, Explainable AI becomes essential for deploying machine learning classifiers. Decision Boundary Mapping (DBM) is a technique for visualizing a classifier's global decision boundary. Despite their relative success, current DBM methods rely on global inverse multidimensional projections that map 2D points back to the input space. The resulting visualizations depend heavily on the chosen projection, distort elements of the structure and distribution of the input data, and contain limited classifier\u2010specific information, often leading to inaccurate boundaries. We propose Sampling\u2010based Precise Neighbourhood Estimation (SPINE), a novel DBM approach for differentiable classifiers that is projection\u2010agnostic and does not depend on inverse projections. SPINE generates additional high\u2010dimensional points near the decision boundary using counterfactuals and variational autoencoders, then applies local interpolation to build visual representations. Our results show that this strategy better captures the structure and distribution of the data and the classifier's predictions in boundary regions than the state\u2010of\u2010the\u2010art in DBMs, yielding higher\u2010resolution and more accurate representations of classifier behaviour. Beyond standard DBM representations, SPINE enables additional exploratory analyses, such as counterfactual feature changes and path visualizations, supporting richer insights into model decisions.<\/jats:p>","DOI":"10.1111\/cgf.70481","type":"journal-article","created":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T10:34:52Z","timestamp":1781174092000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["SPINE: VAE\u2010driven Counterfactuals for Decision Boundary Maps"],"prefix":"10.1111","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-8609-2254","authenticated-orcid":false,"given":"I.M.","family":"Bloemen","sequence":"first","affiliation":[{"name":"Eindhoven University of Technology  The Netherlands"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9296-3693","authenticated-orcid":false,"given":"V.","family":"Prasad","sequence":"additional","affiliation":[{"name":"Eindhoven University of Technology  The Netherlands"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2316-760X","authenticated-orcid":false,"given":"F. V.","family":"Paulovich","sequence":"additional","affiliation":[{"name":"Eindhoven University of Technology  The Netherlands"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,6,11]]},"reference":[{"key":"e_1_2_8_2_2","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2018.2870052"},{"key":"e_1_2_8_3_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.inffus.2019.12.012"},{"key":"e_1_2_8_4_2","doi-asserted-by":"publisher","DOI":"10.1145\/361002.361007"},{"key":"e_1_2_8_5_2","doi-asserted-by":"publisher","DOI":"10.2312\/eurova.20241112"},{"key":"e_1_2_8_6_2","unstructured":"CortezP. CerdeiraA. AlmeidaF. MatosT. ReisJ.:Wine quality data set 2009. UCI Machine Learning Repository. doi:https:\/\/doi.org\/10.24432\/C56S3T. 7"},{"key":"e_1_2_8_7_2","doi-asserted-by":"publisher","DOI":"10.1023\/A:1022664626993"},{"key":"e_1_2_8_8_2","doi-asserted-by":"publisher","DOI":"10.1109\/TSMC.1976.5408784"},{"key":"e_1_2_8_9_2","doi-asserted-by":"publisher","DOI":"10.3389\/fmed.2023.1180773"},{"key":"e_1_2_8_10_2","doi-asserted-by":"publisher","DOI":"10.1109\/TVCG.2021.3125576"},{"key":"e_1_2_8_11_2","doi-asserted-by":"publisher","DOI":"10.5220\/0010184800270037"},{"key":"e_1_2_8_12_2","doi-asserted-by":"publisher","DOI":"10.1109\/TVCG.2019.2944182"},{"key":"e_1_2_8_13_2","doi-asserted-by":"publisher","DOI":"10.5220\/0007260800280038"},{"key":"e_1_2_8_14_2","unstructured":"FeinmanR. CurtinR. R. ShintreS. GardnerA. B.:Detecting adversarial samples from artifacts 2017. URL:https:\/\/arxiv.org\/abs\/1703.00410. 7"},{"key":"e_1_2_8_15_2","unstructured":"GoodfellowI. J. ShlensJ. SzegedyC.:Explaining and harnessing adversarial examples 2015. URL:https:\/\/arxiv.org\/abs\/1412.6572. 8"},{"key":"e_1_2_8_16_2","doi-asserted-by":"crossref","unstructured":"HamelL.: Visualization of support vector machines with un-supervised learning. In2006 IEEE Symposium on Computational Intelligence and Bioinformatics and Computational Biology(2006) pp.1\u20138. doi:10.1109\/CIBCB.2006.330984. 2","DOI":"10.1109\/CIBCB.2006.330984"},{"key":"e_1_2_8_17_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-43427-3_39"},{"key":"e_1_2_8_18_2","doi-asserted-by":"publisher","DOI":"10.1109\/TVCG.2011.220"},{"key":"e_1_2_8_19_2","unstructured":"KingmaD. P. BaJ.: Adam: A method for stochastic optimization. InProc. International Conference on Learning Representations (ICLR)(2015). URL:https:\/\/arxiv.org\/abs\/1412.6980. 5"},{"key":"e_1_2_8_20_2","doi-asserted-by":"publisher","DOI":"10.1214\/aoms\/1177729694"},{"key":"e_1_2_8_21_2","doi-asserted-by":"publisher","DOI":"10.1561\/2200000056"},{"key":"e_1_2_8_22_2","unstructured":"LeCunY.:The MNIST database of handwritten digits 1998. doi:https:\/\/doi.org\/10.24432\/C53K8Q. 6 7"},{"key":"e_1_2_8_23_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.patrec.2005.06.002"},{"key":"e_1_2_8_24_2","doi-asserted-by":"publisher","DOI":"10.1111\/cgf.13672"},{"key":"e_1_2_8_25_2","doi-asserted-by":"publisher","DOI":"10.1111\/cgf.15109"},{"key":"e_1_2_8_26_2","doi-asserted-by":"publisher","DOI":"10.21105\/joss.00861"},{"key":"e_1_2_8_27_2","doi-asserted-by":"publisher","DOI":"10.1109\/TSMC.1987.289362"},{"key":"e_1_2_8_28_2","unstructured":"MartinsR. M. MinghimR. TeleaA. C. et al.: Explaining neighborhood preservation for multidimensional projections. InProc. Computer Graphics and Visual Computing (CGVC)(2015) SciTePress pp.7\u201314. 10"},{"key":"e_1_2_8_29_2","doi-asserted-by":"publisher","DOI":"10.1145\/3351095.3372850"},{"key":"e_1_2_8_30_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.cag.2024.104093"},{"key":"e_1_2_8_31_2","doi-asserted-by":"publisher","DOI":"10.1007\/s10618-013-0342-x"},{"key":"e_1_2_8_32_2","doi-asserted-by":"publisher","DOI":"10.1109\/TVCG.2018.2846735"},{"key":"e_1_2_8_33_2","doi-asserted-by":"publisher","DOI":"10.5220\/0010896200003124"},{"key":"e_1_2_8_34_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.chaos.2023.113130"},{"key":"e_1_2_8_35_2","doi-asserted-by":"publisher","DOI":"10.1109\/TVCG.2023.3301722"},{"key":"e_1_2_8_36_2","unstructured":"PotykaN. YinX. ToniF.:Towards a theory of faithfulness: Faithful explanations of differentiable classifiers over continuous data 2022. URL:https:\/\/arxiv.org\/abs\/2205.09620 arXiv:2205.09620. 3"},{"key":"e_1_2_8_37_2","doi-asserted-by":"publisher","DOI":"10.3390\/info10090280"},{"key":"e_1_2_8_38_2","doi-asserted-by":"publisher","DOI":"10.1109\/TVCG.2016.2598838"},{"key":"e_1_2_8_39_2","doi-asserted-by":"crossref","unstructured":"RodriguesF. C. M. HirataR. TeleaA. C.: Image-based visualization of classifier decision boundaries. InProceedings of the 31st SIBGRAPI Conference on Graphics Patterns and Images(2018) pp.353\u2013360. doi:10.1109\/SIBGRAPI.2018.00052. 2 3 6","DOI":"10.1109\/SIBGRAPI.2018.00052"},{"key":"e_1_2_8_40_2","doi-asserted-by":"publisher","DOI":"10.1016\/0377-0427(87)90125-7"},{"key":"e_1_2_8_41_2","doi-asserted-by":"publisher","DOI":"10.1007\/s11063-014-9394-1"},{"key":"e_1_2_8_42_2","doi-asserted-by":"publisher","DOI":"10.1111\/cgf.14650"},{"key":"e_1_2_8_43_2","unstructured":"SchulzA. HinderF. HammerB.: DeepView: visualizing classification boundaries of deep neural networks as scatter plots using discriminative dimensionality reduction. InProceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence(2021) IJCAI'20. 1 2 3 5 6"},{"key":"e_1_2_8_44_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2023.110273"},{"key":"e_1_2_8_45_2","unstructured":"TanayT. GriffinL.:A boundary tilting persepective on the phenomenon of adversarial examples 2016. URL:https:\/\/arxiv.org\/abs\/1608.07690 arXiv:1608.07690. 7"},{"key":"e_1_2_8_46_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102470"},{"key":"e_1_2_8_47_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-86520-7_40"},{"key":"e_1_2_8_48_2","first-page":"451","article-title":"Information retrieval perspective to nonlinear dimensionality reduction for data visualization","volume":"11","author":"Venna J.","year":"2010","journal-title":"Journal of Machine Learning Research"},{"key":"e_1_2_8_49_2","unstructured":"WolbergW. MangasarianO. StreetN. StreetW.:Breast Cancer Wisconsin (Diagnostic). UCI Machine Learning Repository 1993. 7"},{"key":"e_1_2_8_50_2","doi-asserted-by":"publisher","DOI":"10.3390\/a16090438"},{"key":"e_1_2_8_51_2","doi-asserted-by":"publisher","DOI":"10.5220\/0012352300003660"},{"key":"e_1_2_8_52_2","unstructured":"XiaoH. RasulK. VollgrafR.:Fashion-MNIST: a novel image dataset for benchmarking machine learning algorithms 2017. URL:https:\/\/arxiv.org\/abs\/1708.07747 arXiv:1708.07747. 7"}],"container-title":["Computer Graphics Forum"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/cgf.70481","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1111\/cgf.70481","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/cgf.70481","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T10:35:01Z","timestamp":1781174101000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1111\/cgf.70481"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,11]]},"references-count":51,"alternative-id":["10.1111\/cgf.70481"],"URL":"https:\/\/doi.org\/10.1111\/cgf.70481","archive":["Portico"],"relation":{},"ISSN":["0167-7055","1467-8659"],"issn-type":[{"value":"0167-7055","type":"print"},{"value":"1467-8659","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,11]]},"assertion":[{"value":"2026-06-11","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70481"}}