{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,2]],"date-time":"2026-04-02T02:50:34Z","timestamp":1775098234389,"version":"3.50.1"},"reference-count":28,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2023,3,1]],"date-time":"2023-03-01T00:00:00Z","timestamp":1677628800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,4,18]],"date-time":"2023-04-18T00:00:00Z","timestamp":1681776000000},"content-version":"vor","delay-in-days":48,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/100008398","name":"Villum Fonden","doi-asserted-by":"publisher","award":["Deep Learning and Remote Sensing for Unlocking Global Ecosystem Resource Dynamics (DeReEco)"],"award-info":[{"award-number":["Deep Learning and Remote Sensing for Unlocking Global Ecosystem Resource Dynamics (DeReEco)"]}],"id":[{"id":"10.13039\/100008398","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Deutsches Forschungszentrum f\u00fcr K\u00fcnstliche Intelligenz GmbH (DFKI)"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["K\u00fcnstl Intell"],"published-print":{"date-parts":[[2023,3]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>When training deep learning models for least-squares regression, we cannot expect that the training error residuals of the final model, selected after a fixed training time or based on performance on a hold-out data set, sum to zero. This can introduce a systematic error that accumulates if we are interested in the total aggregated performance over many data points (e.g., the sum of the residuals on previously unseen data). We suggest adjusting the bias of the machine learning model after training as a default post-processing step, which efficiently solves the problem. The severeness of the error accumulation and the effectiveness of the bias correction are demonstrated in exemplary experiments.<\/jats:p>","DOI":"10.1007\/s13218-023-00801-0","type":"journal-article","created":{"date-parts":[[2023,4,18]],"date-time":"2023-04-18T12:02:54Z","timestamp":1681819374000},"page":"33-40","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":6,"title":["Remember to Correct the Bias When Using Deep Learning for Regression!"],"prefix":"10.1007","volume":"37","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2868-0856","authenticated-orcid":false,"given":"Christian","family":"Igel","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0240-1559","authenticated-orcid":false,"given":"Stefan","family":"Oehmcke","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2023,4,18]]},"reference":[{"issue":"1","key":"801_CR1","doi-asserted-by":"publisher","first-page":"49","DOI":"10.1139\/x72-009","volume":"2","author":"G Baskerville","year":"1972","unstructured":"Baskerville G (1972) Use of logarithmic regression in the estimation of plant biomass. Can J For Res 2(1):49\u201353","journal-title":"Can J For Res"},{"key":"801_CR2","doi-asserted-by":"crossref","DOI":"10.1093\/oso\/9780198538493.001.0001","volume-title":"Neural networks for pattern recognition","author":"CM Bishop","year":"1995","unstructured":"Bishop CM (1995) Neural networks for pattern recognition. Oxford University Press, Oxford"},{"key":"801_CR3","doi-asserted-by":"publisher","first-page":"78","DOI":"10.1038\/s41586-020-2824-5","volume":"587","author":"M Brandt","year":"2020","unstructured":"Brandt M, Tucker CJ, Kariryaa A, Rasmussen K, Abel C, Small J, Chave J, Rasmussen LV, Hiernaux P, Diouf AA, Kergoat L, Mertz O, Igel C, Gieseke F, Sch\u00f6ning J, Li S, Melocik K, Meyer J, Sinno S, Romero E, Glennie E, Montagu A, Dendoncker M, Fensholt R (2020) An unexpectedly large count of trees in the western Sahara and Sahel. Nature 587:78\u201382","journal-title":"Nature"},{"issue":"3","key":"801_CR4","doi-asserted-by":"publisher","first-page":"1379","DOI":"10.1021\/ci0504014","volume":"46","author":"P Bruneau","year":"2006","unstructured":"Bruneau P, McElroy NR (2006) logd$$_7.4$$ modeling using Bayesian regularized neural networks. Assessment and correction of the errors of prediction. J Chem Inf Model 46(3):1379\u20131387","journal-title":"J Chem Inf Model"},{"key":"801_CR5","unstructured":"Dua D, Graff C (2017) UCI machine learning repository. University of California, Irvine, School of Information and Computer Sciences. http:\/\/archive.ics.uci.edu\/ml"},{"key":"801_CR6","doi-asserted-by":"publisher","unstructured":"Duncanson L, Armston J, Disney M, Avitabile V, Barbier N, Calders K, Carter S, Chave J, Herold M, MacBean N, McRoberts R, Minor D, Paul K, R\u00e9jou-M\u00e9chain M, Roxburgh S, Williams M, Albinet C, Baker T, Bartholomeus H, Bastin JF, Coomes D, Crowther T, Davies S, de Bruin S, De Kauwe M, Domke G, Dubayah R, Falkowski M, Fatoyinbo L, Goetz S, Jantz P, Jonckheere I, Jucker T, Kay H, Kellner J, Labriere N, Lucas R, Mitchard E, Morsdorf F, Naesset E, Park T, Phillips O, Ploton P, Puliti S, Quegan S, Saatchi S, Schaaf C, Schepaschenko D, Scipal K, Stovall A, Thiel C, Wulder MA, Camacho F, Nickeson J, Rom\u00e1n M, Margolis H (2021) Aboveground woody biomass product validation good practices protocol. Version 1.0. In: Duncanson L, Disney M, Armston J, Nickeson J, Minor D, Camacho F (eds) Good practices for satellite derived land product validation. Committee on Earth Observation Satellites, Land Product Validation Subgroup (WGCV\/CEOS). https:\/\/doi.org\/10.5067\/doc\/ceoswgcv\/lpv\/agb.001","DOI":"10.5067\/doc\/ceoswgcv\/lpv\/agb.001"},{"key":"801_CR7","unstructured":"Espejo A, Federici S, Green C, Amuchastegui N, d\u2019Annunzio R, Balzter H, Bholanath P, Brack C, Brewer C, Birigazzi L, Cabrera E, Carter S, Chand N, Donoghue D, Eggleston S, Fitzgerald N, Foody G, Galindo G, Goeking S, Grassi G, Held A, Herold M (2020) Integration of remote-sensing and ground-based observations for estimation of emissions and removals of greenhouse gases in forests: methods and guidance from the Global Forest Observations Initiative, Edition 3.0. UN Food and Agriculture Organization"},{"issue":"2","key":"801_CR8","doi-asserted-by":"publisher","first-page":"155","DOI":"10.2307\/2983663","volume":"7","author":"D Finney","year":"1941","unstructured":"Finney D (1941) On the distribution of a variate whose logarithm is normally distributed. Suppl J R Stat Soc 7(2):155\u2013161","journal-title":"Suppl J R Stat Soc"},{"key":"801_CR9","unstructured":"Guo C, Pleiss G, Sun Y, Weinberger KQ (2017) On calibration of modern neural networks. In: International conference on machine learning (ICML), PMLR 70, pp 1321\u20131330"},{"key":"801_CR10","doi-asserted-by":"crossref","unstructured":"Hu J, Shen L, Sun G (2018) Squeeze-and-excitation networks. In: Computer vision and pattern recognition (CVPR). IEEE, pp 7132\u20137141","DOI":"10.1109\/CVPR.2018.00745"},{"issue":"1","key":"801_CR11","doi-asserted-by":"publisher","first-page":"177","DOI":"10.1111\/gcb.13388","volume":"23","author":"T Jucker","year":"2017","unstructured":"Jucker T, Caspersen J, Chave J, Antin C, Barbier N, Bongers F, Dalponte M, van Ewijk KY, Forrester DI, Haeni M, Higgins SI, Holdaway RJ, Iida Z, Lorime C, Marshall PL, Momo S, Moncrieff GR, Ploton P, Poorter L, Rahman KA, Schlund M, Sonk\u00e9 B, Sterck FJ, Trugman AT, Usoltsev VA, Vanderwel MC, Waldner P, Wedeux BMM, Wirth C, W\u00f6ll H, Woods M, Xiang W, Zimmermann NE, Coomes DA (2017) Allometric equations for integrating remote sensing imagery into forest monitoring programmes. Glob Change Biol 23(1):177\u2013190","journal-title":"Glob Change Biol"},{"issue":"6","key":"801_CR12","doi-asserted-by":"publisher","first-page":"4783","DOI":"10.3906\/elk-1807-87","volume":"27","author":"H Kaya","year":"2019","unstructured":"Kaya H, T\u00fcfekci P, Uzun E (2019) Predicting CO and NOx emissions from gas turbines: novel data and a benchmark PEMS. Turk J Electr Eng Comput Sci 27(6):4783\u20134796","journal-title":"Turk J Electr Eng Comput Sci"},{"key":"801_CR13","unstructured":"Kinga DP, Ba JL (2015) Adam: a method for stochastic optimization. In: International conference on learning representations (ICLR)"},{"key":"801_CR14","doi-asserted-by":"publisher","DOI":"10.1016\/j.ecolind.2021.108081","volume":"130","author":"TG Morais","year":"2021","unstructured":"Morais TG, Teixeira RF, Figueiredo M, Domingos T (2021) The use of machine learning methods to estimate aboveground biomass of grasslands: a review. Ecol Indic 130:108081","journal-title":"Ecol Indic"},{"key":"801_CR15","doi-asserted-by":"crossref","unstructured":"Mugabowindekwe M, Brandt M, Chave J, Reiner F, Skole D, Kariryaa A, Igel C, Hiernaux P, Ciais P, Mertz O, Tong X, Li S, Rwanyiziri G, Dushimiyimana T, Ndoli A, Uwizeyimana V, Lilles\u00f8 JP, Gieseke F, Tucker C, Saatchi SS, Fensholt R (2023) Nation-wide mapping of tree-level aboveground carbon stocks in Rwanda. Nat Clim Change 13:91\u201397","DOI":"10.1038\/s41558-022-01544-w"},{"key":"801_CR16","volume":"93","author":"E N\u0153sset","year":"2020","unstructured":"N\u0153sset E, McRoberts RE, Pekkarinen A, Saatchi S, Santoro M, Trier \u00d8ivind D, Zahabu E, Gobakken T (2020) Use of local and global maps of forest canopy height and aboveground biomass to enhance local estimates of biomass in miombo woodlands in Tanzania. Int J Appl Earth Observ Geoinf 93:102138","journal-title":"Int J Appl Earth Observ Geoinf"},{"key":"801_CR17","doi-asserted-by":"crossref","unstructured":"Oehmcke S, Li L, Revenga J, Nord-Larsen T, Trepekli K, Gieseke F, Igel C (2022) Deep learning based 3D point cloud regression for estimating forest biomass. In: International conference on advances in geographic information systems (SIGSPATIAL). ACM, pp 38:1\u201338:4","DOI":"10.1145\/3557915.3561471"},{"key":"801_CR18","unstructured":"Paszke A, Gross S, Massa F, Lerer A, Bradbury J, Chanan G, Killeen T, Lin Z, Gimelshein N, Antiga L, Desmaison A, Kopf A, Yang E, DeVito Z, Raison M, Tejani A, Chilamkurthy S, Steiner B, Fang L, Bai J, Chintala S (2019) PyTorch: an imperative style, high-performance deep learning library. In: Wallach H, Larochelle H, Beygelzimer A, d\u2019Alch\u00e9-Buc F, Fox E, Garnett R (eds) Advances in neural information processing systems (NeurIPS), pp 8024\u20138035"},{"issue":"85","key":"801_CR19","first-page":"2825","volume":"12","author":"F Pedregosa","year":"2011","unstructured":"Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V, Vanderplas J, Passos A, Cournapeau D, Brucher M, Perrot M, Duchesnay \u00c9douard (2011) Scikit-learn: machine learning in Python. J Mach Learn Res 12(85):2825\u20132830","journal-title":"J Mach Learn Res"},{"key":"801_CR20","doi-asserted-by":"crossref","unstructured":"Persello C, Wegner JD, Hansch R, Tuia D, Ghamisi P, Koeva M, Camps-Valls G (2022) Deep learning and earth observation to support the sustainable development goals: Current approaches, open challenges, and future opportunities. IEEE Geosci Remote Sens Mag 10(2):172\u2013200","DOI":"10.1109\/MGRS.2021.3136100"},{"key":"801_CR21","doi-asserted-by":"publisher","first-page":"53","DOI":"10.1007\/978-3-642-35289-8_5","volume-title":"Neural networks: tricks of the trade","author":"L Prechelt","year":"2012","unstructured":"Prechelt L (2012) Early stopping\u2014but when? In: Montavon G, Orr GB, M\u00fcller KR (eds) Neural networks: tricks of the trade, 2nd edn. Springer, Berlin, pp 53\u201367","edition":"2"},{"issue":"6","key":"801_CR22","doi-asserted-by":"publisher","first-page":"2401","DOI":"10.1021\/ci700197x","volume":"47","author":"SL Rodgers","year":"2007","unstructured":"Rodgers SL, Davis AM, Tomkinson NP, van de Waterbeemd H (2007) QSAR modeling using automatically updating correction libraries: application to a human plasma protein binding model. J Chem Inf Model 47(6):2401\u20132407","journal-title":"J Chem Inf Model"},{"key":"801_CR23","unstructured":"Song H, Diethe T, Kull M, Flach P (2019) Distribution calibration for regression. In: International conference on machine learning (ICML), PMLR, pp 5897\u20135906"},{"key":"801_CR24","unstructured":"Tan M, Le Q (2019) EfficientNet: rethinking model scaling for convolutional neural networks. In: International conference on machine learning (ICML), PMLR, pp 6105\u20136114"},{"key":"801_CR25","doi-asserted-by":"crossref","unstructured":"Tan M, Chen B, Pang R, Vasudevan V, Sandler M, Howard A, Le QV (2019) MnasNet: platform-aware neural architecture search for mobile. In: Computer vision and pattern recognition (CVPR). IEEE, pp 2820\u20132828","DOI":"10.1109\/CVPR.2019.00293"},{"key":"801_CR26","doi-asserted-by":"publisher","DOI":"10.1016\/j.rse.2020.111716","volume":"241","author":"Q Yuan","year":"2020","unstructured":"Yuan Q, Shen H, Li T, Li Z, Li S, Jiang Y, Xu H, Tan W, Yang Q, Wang J, Gao J, Zhang L (2020) Deep learning in environmental remote sensing: achievements and challenges. Remote Sens Environ 241:111716","journal-title":"Remote Sens Environ"},{"key":"801_CR27","doi-asserted-by":"publisher","unstructured":"Zeng W, Tang S (2011) Bias correction in logarithmic regression and comparison with weighted regression for nonlinear models. Nat Preced. https:\/\/doi.org\/10.1038\/npre.2011.6708.1","DOI":"10.1038\/npre.2011.6708.1"},{"key":"801_CR28","doi-asserted-by":"crossref","unstructured":"Zhang Y, Ma J, Liang S, Li X, Li M (2020) An evaluation of eight machine learning regression algorithms for forest aboveground biomass estimation from multiple satellite data products. Remote Sens 12(24):4015","DOI":"10.3390\/rs12244015"}],"container-title":["KI - K\u00fcnstliche Intelligenz"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s13218-023-00801-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s13218-023-00801-0\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s13218-023-00801-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,12,10]],"date-time":"2023-12-10T21:23:25Z","timestamp":1702243405000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s13218-023-00801-0"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,3]]},"references-count":28,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2023,3]]}},"alternative-id":["801"],"URL":"https:\/\/doi.org\/10.1007\/s13218-023-00801-0","relation":{},"ISSN":["0933-1875","1610-1987"],"issn-type":[{"value":"0933-1875","type":"print"},{"value":"1610-1987","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,3]]},"assertion":[{"value":"12 July 2022","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"16 November 2022","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"18 April 2023","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}}]}}