{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,20]],"date-time":"2026-11-20T04:25:42Z","timestamp":1795148742563,"version":"4.10.3"},"reference-count":49,"publisher":"Association for Computing Machinery (ACM)","issue":"9","funder":[{"DOI":"10.13039\/100000001","name":"US National Science Foundation","doi-asserted-by":"crossref","award":["IIS-2106913, SCH-2500341, CCF-2217071, BIO-2313865, and CNS-2213700"],"award-info":[{"award-number":["IIS-2106913, SCH-2500341, CCF-2217071, BIO-2313865, and CNS-2213700"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Knowl. Discov. Data"],"published-print":{"date-parts":[[2025,11,30]]},"abstract":"<jats:p>\n            Generalized Additive Models (GAMs) have long been a powerful white-box tool for the intelligible analysis of tabular data, revealing the influence of each feature on the model predictions. Despite the success of Neural Networks (NNs) in various domains, their application as NN-based GAMs in tabular data analysis remains suboptimal compared to tree-based ones, and the opacity of encoders in NN-GAMs also prevents users from understanding how networks learn the functions. In this work, we propose a new deep tabular learning method, termed Prototypical Neural Additive Model (ProtoNAM), which introduces prototypes into NNs in the framework of GAMs. With the introduced prototype-based feature activation, ProtoNAM can flexibly model the irregular mapping from tabular features to the outputs while maintaining the explainability of the final prediction. We also propose a gradient-boosting inspired hierarchical shape function modeling method, facilitating the discovery of complex feature patterns and bringing transparency into the learning process of each network layer. Our empirical evaluations demonstrate that ProtoNAM outperforms all existing NN-based GAMs, while providing additional insights into the shape function learned for each feature. The source code of ProtoNAM is available at\n            <jats:ext-link xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" ext-link-type=\"uri\" xlink:href=\"https:\/\/github.com\/Teddy-XiongGZ\/ProtoNAM\">https:\/\/github.com\/Teddy-XiongGZ\/ProtoNAM<\/jats:ext-link>\n            .\n          <\/jats:p>","DOI":"10.1145\/3766072","type":"journal-article","created":{"date-parts":[[2025,9,5]],"date-time":"2025-09-05T15:11:28Z","timestamp":1757085088000},"page":"1-25","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["ProtoNAM: Prototypical Neural Additive Models for Interpretable Deep Tabular Learning"],"prefix":"10.1145","volume":"19","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8049-5298","authenticated-orcid":false,"given":"Guangzhi","family":"Xiong","sequence":"first","affiliation":[{"name":"Department of Computer Science, University of Virginia, Charlottesville, Virginia, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2650-0612","authenticated-orcid":false,"given":"Sanchit","family":"Sinha","sequence":"additional","affiliation":[{"name":"Department of Computer Science, University of Virginia, Charlottesville, Virginia, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9723-3246","authenticated-orcid":false,"given":"Aidong","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Computer Science, University of Virginia, Charlottesville, Virginia, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2025,10,7]]},"reference":[{"key":"e_1_3_1_2_2","first-page":"4699","article-title":"Neural additive models: Interpretable machine learning with neural nets","volume":"34","author":"Agarwal Rishabh","year":"2021","unstructured":"Rishabh Agarwal, Levi Melnick, Nicholas Frosst, Xuezhou Zhang, Ben Lengerich, Rich Caruana, and Geoffrey E. Hinton. 2021. Neural additive models: Interpretable machine learning with neural nets. In Proceedings of the Advances in Neural Information Processing Systems, Vol. 34, 4699\u20134711.","journal-title":"Proceedings of the Advances in Neural Information Processing Systems"},{"key":"e_1_3_1_3_2","unstructured":"Dara Bahri Heinrich Jiang Yi Tay and Donald Metzler. 2021. SCARF: Self-supervised contrastive learning using random feature corruption. arXiv:2106.15147. Retrieved from https:\/\/arxiv.org\/abs\/2106.15147"},{"key":"e_1_3_1_4_2","doi-asserted-by":"publisher","DOI":"10.1214\/11-AOAS495"},{"key":"e_1_3_1_5_2","unstructured":"Catherine Blake. 1998. UCI Repository of Machine Learning Databases. Retrieved from http:\/\/www.ics.uci.edu\/\u223cmlearn\/MLRepository.html"},{"key":"e_1_3_1_6_2","volume-title":"Proceedings of the 10th International Conference on Learning Representations (ICLR \u201922), Virtual Event","author":"Chang Chun-Hao","year":"2022","unstructured":"Chun-Hao Chang, Rich Caruana, and Anna Goldenberg. 2022. NODE-GAM: Neural generalized additive model for interpretable deep learning. In Proceedings of the 10th International Conference on Learning Representations (ICLR \u201922), Virtual Event. OpenReview.net. Retrieved from https:\/\/openreview.net\/forum?id=g8NJR6fCCl8"},{"key":"e_1_3_1_7_2","doi-asserted-by":"publisher","DOI":"10.1145\/3447548.3467453"},{"key":"e_1_3_1_8_2","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v36i4.20309"},{"key":"e_1_3_1_9_2","doi-asserted-by":"publisher","DOI":"10.1145\/2939672.2939785"},{"key":"e_1_3_1_10_2","unstructured":"Anubrata Das Chitrank Gupta Venelin Kovatchev Matthew Lease and Junyi Jessy Li. 2022. ProtoTEx: Explaining model decisions with prototype tensors. arXiv:2204.05426. Retrieved from https:\/\/arxiv.org\/abs\/2204.05426"},{"key":"e_1_3_1_11_2","unstructured":"Jacob Devlin. 2018. BERT: Pre-training of deep bidirectional transformers for language understanding. arXiv:1810.04805. Retrieved from https:\/\/arxiv.org\/abs\/1810.04805"},{"key":"e_1_3_1_12_2","doi-asserted-by":"publisher","DOI":"10.1214\/aos\/1013203451"},{"key":"e_1_3_1_13_2","first-page":"507","volume-title":"Proceedings of the Advances in Neural Information Processing Systems","author":"Grinsztajn Leo","year":"2022","unstructured":"Leo Grinsztajn, Edouard Oyallon, and Gael Varoquaux. 2022. Why do tree-based models still outperform deep learning on typical tabular data? In Proceedings of the Advances in Neural Information Processing Systems. S. Koyejo, S. Mohamed, A. Agarwal, D. Belgrave, K. Cho, and A. Oh (Eds.), Vol. 3, Curran Associates, Inc., 507\u2013520. Retrieved from https:\/\/proceedings.neurips.cc\/paper_files\/paper\/2022\/file\/0378c7692da36807bdec87ab043cdadc-Paper-Datasets_and_Benchmarks.pdf"},{"key":"e_1_3_1_14_2","first-page":"32","volume-title":"Proceedings of the AAAI Conference on Human Computation and Crowdsourcing","volume":"7","author":"Hase Peter","year":"2019","unstructured":"Peter Hase, Chaofan Chen, Oscar Li, and Cynthia Rudin. 2019. Interpretable image recognition with hierarchical prototypes. In Proceedings of the AAAI Conference on Human Computation and Crowdsourcing, Vol. 7, 32\u201340."},{"issue":"3","key":"e_1_3_1_15_2","doi-asserted-by":"crossref","first-page":"187","DOI":"10.1177\/096228029500400302","article-title":"Generalized additive models for medical research","volume":"4","author":"Hastie Trevor","year":"1995","unstructured":"Trevor Hastie and Robert Tibshirani. 1995. Generalized additive models for medical research. Statistical Methods in Medical Research 4, 3 (1995), 187\u2013196.","journal-title":"Statistical Methods in Medical Research"},{"key":"e_1_3_1_16_2","doi-asserted-by":"publisher","DOI":"10.1201\/9780203738535-7"},{"key":"e_1_3_1_17_2","first-page":"46","volume-title":"Proceedings of the Machine Learning for Healthcare Conference","author":"Hegselmann Stefan","year":"2020","unstructured":"Stefan Hegselmann, Thomas Volkert, Hendrik Ohlenburg, Antje Gottschalk, Martin Dugas, and Christian Ertmer. 2020. An evaluation of the doctor-interpretability of generalized additive models with interactions. In Proceedings of the Machine Learning for Healthcare Conference. PMLR, 46\u201379."},{"key":"e_1_3_1_18_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-024-08328-6"},{"key":"e_1_3_1_19_2","first-page":"153","volume-title":"Mathematics of Public Health: Proceedings of the Seminar on the Mathematical Modelling of COVID-19","author":"Izadi Farzali","year":"2021","unstructured":"Farzali Izadi. 2021. Generalized additive models to capture the death rates in Canada COVID-19. In Mathematics of Public Health: Proceedings of the Seminar on the Mathematical Modelling of COVID-19. Springer, 153\u2013171."},{"key":"e_1_3_1_20_2","doi-asserted-by":"publisher","DOI":"10.1038\/sdata.2016.35"},{"key":"e_1_3_1_21_2","doi-asserted-by":"publisher","DOI":"10.3390\/s23031713"},{"key":"e_1_3_1_22_2","doi-asserted-by":"publisher","unstructured":"Salman Khan Hossein Rahmani Syed Afaq Ali Shah Mohammed Bennamoun. 2018. A Guide to Convolutional Neural Networks for Computer Vision (1st ed.). Springer Cham. DOI: 10.1007\/978-3-031-01821-3","DOI":"10.1007\/978-3-031-01821-3"},{"key":"e_1_3_1_23_2","doi-asserted-by":"publisher","DOI":"10.1007\/s11042-022-13428-4"},{"key":"e_1_3_1_24_2","article-title":"The bayesian case model: A generative approach for case-based reasoning and prototype classification","volume":"27","author":"Kim Been","year":"2014","unstructured":"Been Kim, Cynthia Rudin, and Julie A. Shah. 2014. The bayesian case model: A generative approach for case-based reasoning and prototype classification. In Proceedings of the Advances in Neural Information Processing Systems, Vol. 27.","journal-title":"Proceedings of the Advances in Neural Information Processing Systems"},{"key":"e_1_3_1_25_2","doi-asserted-by":"publisher","DOI":"10.1145\/3065386"},{"key":"e_1_3_1_26_2","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.11771"},{"key":"e_1_3_1_27_2","doi-asserted-by":"publisher","DOI":"10.1145\/2339530.2339556"},{"key":"e_1_3_1_28_2","article-title":"A unified approach to interpreting model predictions","volume":"30","author":"Lundberg Scott M.","year":"2017","unstructured":"Scott M. Lundberg and Su-In Lee. 2017. A unified approach to interpreting model predictions. In Proceedings of the Advances in Neural Information Processing Systems, Vol. 30.","journal-title":"Proceedings of the Advances in Neural Information Processing Systems"},{"key":"e_1_3_1_29_2","article-title":"When do neural nets outperform boosted trees on tabular data","volume":"36","author":"McElfresh Duncan","year":"2024","unstructured":"Duncan McElfresh, Sujay Khandagale, Jonathan Valverde, C. Vishak Prasad, Ganesh Ramakrishnan, Micah Goldblum, and Colin White. 2024. When do neural nets outperform boosted trees on tabular data? In Proceedings of the Advances in Neural Information Processing Systems, Vol. 36.","journal-title":"Proceedings of the Advances in Neural Information Processing Systems"},{"key":"e_1_3_1_30_2","doi-asserted-by":"publisher","DOI":"10.1145\/3605943"},{"key":"e_1_3_1_31_2","doi-asserted-by":"publisher","DOI":"10.1145\/3292500.3330908"},{"key":"e_1_3_1_32_2","volume-title":"Proceedings of the International Conference on Learning Representations","author":"M\u00fcller Samuel","year":"2022","unstructured":"Samuel M\u00fcller, Noah Hollmann, Sebastian Pineda Arango, Josif Grabocka, and Frank Hutter. 2022. Transformers can do bayesian inference. In Proceedings of the International Conference on Learning Representations. Retrieved from https:\/\/openreview.net\/forum?id=KSugKcbNf9"},{"key":"e_1_3_1_33_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01469"},{"key":"e_1_3_1_34_2","unstructured":"Harsha Nori Samuel Jenkins Paul Koch and Rich Caruana. 2019. InterpretML: A unified framework for machine learning interpretability. arXiv:1909.09223. Retrieved from https:\/\/arxiv.org\/abs\/1909.09223"},{"key":"e_1_3_1_35_2","doi-asserted-by":"publisher","DOI":"10.1016\/S0167-7152(96)00140-X"},{"key":"e_1_3_1_36_2","doi-asserted-by":"publisher","DOI":"10.7717\/peerj.6876"},{"key":"e_1_3_1_37_2","volume-title":"Proceedings of the 8th International Conference on Learning Representations (ICLR \u201920)","author":"Popov Sergei","year":"2020","unstructured":"Sergei Popov, Stanislav Morozov, and Artem Babenko. 2020. Neural oblivious decision ensembles for deep learning on tabular data. In Proceedings of the 8th International Conference on Learning Representations (ICLR \u201920). OpenReview.net. Retrieved from https:\/\/openreview.net\/forum?id=r1eiu2VtwH"},{"key":"e_1_3_1_38_2","first-page":"8414","volume-title":"Proceedings of the Advances in Neural Information Processing Systems","author":"Radenovic Filip","year":"2022","unstructured":"Filip Radenovic, Abhimanyu Dubey, and Dhruv Mahajan. 2022. Neural Basis Models for Interpretability. In Proceedings of the Advances in Neural Information Processing Systems. S. Koyejo, S. Mohamed, A. Agarwal, D. Belgrave, K. Cho, and A. Oh (Eds.), Vol. 35, Curran Associates, Inc., 8414\u20138426. Retrieved from https:\/\/proceedings.neurips.cc\/paper_files\/paper\/2022\/file\/37da88965c016dca016514df0e420c72-Paper-Conference.pdf"},{"key":"e_1_3_1_39_2","unstructured":"S. Ruder. 2017. An overview of multi-task learning in deep neural networks. arXiv:1706.05098. Retrieved from https:\/\/arxiv.org\/abs\/1706.05098"},{"key":"e_1_3_1_40_2","doi-asserted-by":"publisher","DOI":"10.1097\/CCM.0b013e31820a92c6"},{"issue":"3","key":"e_1_3_1_41_2","first-page":"64","article-title":"Generalized additive models in business and economics","volume":"1","author":"Sapra Sunil K.","year":"2013","unstructured":"Sunil K. Sapra. 2013. Generalized additive models in business and economics. International Journal of Advanced Statistics and Probability 1, 3 (2013), 64\u201381.","journal-title":"International Journal of Advanced Statistics and Probability"},{"key":"e_1_3_1_42_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.inffus.2021.11.011"},{"key":"e_1_3_1_43_2","first-page":"503","volume-title":"Proceedings of the 33rd International Joint Conference on Artificial Intelligence","author":"Sinha Sanchit","year":"2024","unstructured":"Sanchit Sinha, Guangzhi Xiong, and Aidong Zhang. 2024. A self-explaining neural architecture for generalizable concept learning. In Proceedings of the 33rd International Joint Conference on Artificial Intelligence, 503\u2013511."},{"key":"e_1_3_1_44_2","article-title":"Prototypical networks for few-shot learning","volume":"30","author":"Snell Jake","year":"2017","unstructured":"Jake Snell, Kevin Swersky, and Richard Zemel. 2017. Prototypical networks for few-shot learning. In Proceedings of the Advances in Neural Information Processing Systems, Vol. 30.","journal-title":"Proceedings of the Advances in Neural Information Processing Systems"},{"key":"e_1_3_1_45_2","unstructured":"Gowthami Somepalli Micah Goldblum Avi Schwarzschild C. Bayan Bruss and Tom Goldstein. 2021. SAINT: Improved neural networks for tabular data via row attention and contrastive pre-training. arXiv:2106.01342. Retrieved from https:\/\/arxiv.org\/abs\/2106.01342"},{"issue":"8","key":"e_1_3_1_46_2","first-page":"btad493","article-title":"ProtoCell4P: An explainable prototype-based neural network for patient classification using single-cell RNA-seq","volume":"39","author":"Xiong Guangzhi","year":"2023","unstructured":"Guangzhi Xiong, Stefan Bekiranov, and Aidong Zhang. 2023. ProtoCell4P: An explainable prototype-based neural network for patient classification using single-cell RNA-seq. Bioinformatics (Oxford, England) 39, 8 (2023), btad493.","journal-title":"Bioinformatics (Oxford, England)"},{"issue":"7","key":"e_1_3_1_47_2","first-page":"btae434","article-title":"DeepGSEA: Explainable deep gene set enrichment analysis for single-cell transcriptomic data","volume":"40","author":"Xiong Guangzhi","year":"2024","unstructured":"Guangzhi Xiong, Nathan J. LeRoy, Stefan Bekiranov, Nathan C. Sheffield, and Aidong Zhang. 2024. DeepGSEA: Explainable deep gene set enrichment analysis for single-cell transcriptomic data. Bioinformatics (Oxford, England) 40, 7 (2024), btae434.","journal-title":"Bioinformatics (Oxford, England)"},{"key":"e_1_3_1_48_2","doi-asserted-by":"publisher","DOI":"10.1109\/MCI.2018.2840738"},{"key":"e_1_3_1_49_2","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i15.29628"},{"key":"e_1_3_1_50_2","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-024-10721-6"}],"container-title":["ACM Transactions on Knowledge Discovery from Data"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3766072","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,7]],"date-time":"2025-10-07T13:57:03Z","timestamp":1759845423000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3766072"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,7]]},"references-count":49,"journal-issue":{"issue":"9","published-print":{"date-parts":[[2025,11,30]]}},"alternative-id":["10.1145\/3766072"],"URL":"https:\/\/doi.org\/10.1145\/3766072","relation":{},"ISSN":["1556-4681","1556-472X"],"issn-type":[{"value":"1556-4681","type":"print"},{"value":"1556-472X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,7]]},"assertion":[{"value":"2025-01-05","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-08-07","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-10-07","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}