{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,28]],"date-time":"2026-07-28T15:03:10Z","timestamp":1785250990951,"version":"3.55.0"},"reference-count":43,"publisher":"Springer Science and Business Media LLC","issue":"4","license":[{"start":{"date-parts":[[2026,5,14]],"date-time":"2026-05-14T00:00:00Z","timestamp":1778716800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,5,14]],"date-time":"2026-05-14T00:00:00Z","timestamp":1778716800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100003790","name":"Hiroshima University","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100003790","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001691","name":"Japan Society for the Promotion of Science","doi-asserted-by":"publisher","award":["JP22K11934"],"award-info":[{"award-number":["JP22K11934"]}],"id":[{"id":"10.13039\/501100001691","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Stat Comput"],"published-print":{"date-parts":[[2026,8]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Gaussian process (GP) regression is often the first choice for prediction with well-calibrated probabilities. However, since GP regression can be very sensitive to outliers, various methods have been proposed to improve robustness, for example, by replacing the Gaussian likelihood with a Student-t distribution. Unfortunately, for predicting count data, like the number of hospitalized patients, such methods are not appropriate. As a solution, we propose to use a latent GP regression model with a negative binomial likelihood that is trained by optimizing a trimmed variational lower bound, where the trimming ratio corresponds to an upper-bound on the number of outliers. We formulate the optimization problem using the sparse variational inference framework, which allows us to derive an efficient training algorithm with provable robustness to outliers. Furthermore, since an unnecessary conservative estimate of the outlier upper-bound can hurt statistical efficiency, we also propose an effective refinement method for tuning the trimming ratio in an automated fashion. Experiments on synthetic, and various real datasets confirm that optimizing the trimmed variational lower bound, rather than the standard variational lower bound, leads to improved predictive uncertainties that are almost unaffected by outliers in the training data. We also generalize and compare our approach to the Bayesian data re-weighting framework and the\n                    <jats:inline-formula>\n                      <jats:alternatives>\n                        <jats:tex-math>$$\\gamma $$<\/jats:tex-math>\n                        <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                          <mml:mi>\u03b3<\/mml:mi>\n                        <\/mml:math>\n                      <\/jats:alternatives>\n                    <\/jats:inline-formula>\n                    -divergence for count data, showing experimentally that the latter is inferior to our proposed methodology.\n                  <\/jats:p>","DOI":"10.1007\/s11222-026-10895-9","type":"journal-article","created":{"date-parts":[[2026,5,14]],"date-time":"2026-05-14T04:25:49Z","timestamp":1778732749000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Robust Variational Gaussian Process Regression for Count Data with the Trimmed Marginal Likelihood"],"prefix":"10.1007","volume":"36","author":[{"given":"Daniel","family":"Andrade","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,5,14]]},"reference":[{"issue":"4","key":"10895_CR1","doi-asserted-by":"publisher","first-page":"920","DOI":"10.1111\/biom.12212","volume":"70","author":"WH Aeberhard","year":"2014","unstructured":"Aeberhard, W.H., Cantoni, E., Heritier, S.: Robust inference in the negative binomial regression model with an application to falls data. Biometrics 70(4), 920\u2013931 (2014)","journal-title":"Biometrics"},{"key":"10895_CR2","volume-title":"An introduction to multivariate statistical analysis","author":"TW Anderson","year":"2003","unstructured":"Anderson, T.W.: An introduction to multivariate statistical analysis, vol. 2. Wiley, New York (2003)"},{"issue":"4","key":"10895_CR3","doi-asserted-by":"publisher","DOI":"10.1002\/sta4.70018","volume":"13","author":"D Andrade","year":"2024","unstructured":"Andrade, D.: Improved variance estimation from trimmed samples. Stat 13(4), e70018 (2024)","journal-title":"Stat"},{"key":"10895_CR4","unstructured":"Andrade, D., Takeda, A.: Robust gaussian process regression with the trimmed marginal likelihood. In: Uncertainty in Artificial Intelligence, PMLR, pp 67\u201376 (2023)"},{"issue":"3","key":"10895_CR5","doi-asserted-by":"publisher","first-page":"64","DOI":"10.1007\/s11222-023-10225-3","volume":"33","author":"M Basson","year":"2023","unstructured":"Basson, M., Louw, T.M., Smith, T.R.: Variational tobit gaussian process regression. Stat. Comput. 33(3), 64 (2023)","journal-title":"Stat. Comput."},{"issue":"3","key":"10895_CR6","doi-asserted-by":"publisher","first-page":"549","DOI":"10.1093\/biomet\/85.3.549","volume":"85","author":"A Basu","year":"1998","unstructured":"Basu, A., Harris, I.R., Hjort, N.L., et al.: Robust and efficient estimation by minimising a density power divergence. Biometrika 85(3), 549\u2013559 (1998)","journal-title":"Biometrika"},{"issue":"1","key":"10895_CR7","doi-asserted-by":"publisher","first-page":"140","DOI":"10.1214\/22-STS864","volume":"38","author":"D Bolin","year":"2023","unstructured":"Bolin, D., Wallin, J.: Local scale invariance and robustness of proper scoring rules. Stat. Sci. 38(1), 140\u2013159 (2023)","journal-title":"Stat. Sci."},{"issue":"27","key":"10895_CR8","doi-asserted-by":"publisher","first-page":"4805","DOI":"10.1002\/sim.6256","volume":"33","author":"R Chambers","year":"2014","unstructured":"Chambers, R., Dreassi, E., Salvati, N.: Disease mapping via negative binomial regression m-quantiles. Stat. Med. 33(27), 4805\u20134824 (2014)","journal-title":"Stat. Med."},{"key":"10895_CR9","unstructured":"Concei\u00e7\u00e3o, K.S., Andrade, M.G., Louzada, F., et\u00a0al.: Characterizations and generalizations of the negative binomial distribution. Computational Statistics pp 1\u201332 (2022)"},{"issue":"4","key":"10895_CR10","doi-asserted-by":"publisher","first-page":"1254","DOI":"10.1111\/j.1541-0420.2009.01191.x","volume":"65","author":"C Czado","year":"2009","unstructured":"Czado, C., Gneiting, T., Held, L.: Predictive model assessment for count data. Biometrics 65(4), 1254\u20131261 (2009)","journal-title":"Biometrics"},{"issue":"4","key":"10895_CR11","doi-asserted-by":"publisher","first-page":"971","DOI":"10.1093\/biomet\/asw053","volume":"103","author":"J Datta","year":"2016","unstructured":"Datta, J., Dunson, D.B.: Bayesian inference on quasi-sparse count data. Biometrika 103(4), 971\u2013983 (2016)","journal-title":"Biometrika"},{"key":"10895_CR12","doi-asserted-by":"publisher","DOI":"10.1007\/978-981-97-8880-4","volume-title":"Minimum Gamma Divergence for Regression and Classification Problems","author":"S Eguchi","year":"2025","unstructured":"Eguchi, S.: Minimum Gamma Divergence for Regression and Classification Problems. Springer (2025)"},{"key":"10895_CR13","unstructured":"Flaxman, S., Wilson, A., Neill, D., et\u00a0al.: Fast kronecker inference in gaussian processes with non-gaussian likelihoods. In: International conference on machine learning, PMLR, pp 607\u2013616 (2015)"},{"issue":"1","key":"10895_CR14","first-page":"1","volume":"19","author":"JH Friedman","year":"1991","unstructured":"Friedman, J.H.: Multivariate adaptive regression splines. Ann. Stat. 19(1), 1\u201367 (1991)","journal-title":"Ann. Stat."},{"issue":"9","key":"10895_CR15","doi-asserted-by":"publisher","first-page":"2053","DOI":"10.1016\/j.jmva.2008.02.004","volume":"99","author":"H Fujisawa","year":"2008","unstructured":"Fujisawa, H., Eguchi, S.: Robust parameter estimation with a small bias against heavy contamination. J. Multivar. Anal. 99(9), 2053\u20132081 (2008)","journal-title":"J. Multivar. Anal."},{"key":"10895_CR16","unstructured":"Futami, F., Sato, I., Sugiyama, M.: Variational inference based on robust divergences. In: International Conference on Artificial Intelligence and Statistics, PMLR, pp 813\u2013822 (2018)"},{"issue":"3","key":"10895_CR17","doi-asserted-by":"publisher","first-page":"1304","DOI":"10.1016\/j.csda.2010.09.019","volume":"55","author":"AM Garay","year":"2011","unstructured":"Garay, A.M., Hashimoto, E.M., Ortega, E.M., et al.: On estimation and influence diagnostics for zero-inflated negative binomial regression models. Computational Statistics & Data Analysis 55(3), 1304\u20131318 (2011)","journal-title":"Computational Statistics & Data Analysis"},{"key":"10895_CR18","unstructured":"Gardner, J., Pleiss, G., Weinberger, K.Q., et\u00a0al.: Gpytorch: Blackbox matrix-matrix gaussian process inference with gpu acceleration. Advances in neural information processing systems 31 (2018)"},{"issue":"477","key":"10895_CR19","doi-asserted-by":"publisher","first-page":"359","DOI":"10.1198\/016214506000001437","volume":"102","author":"T Gneiting","year":"2007","unstructured":"Gneiting, T., Raftery, A.E.: Strictly proper scoring rules, prediction, and estimation. J. Am. Stat. Assoc. 102(477), 359\u2013378 (2007)","journal-title":"J. Am. Stat. Assoc."},{"issue":"2","key":"10895_CR20","doi-asserted-by":"publisher","first-page":"243","DOI":"10.1111\/j.1467-9868.2007.00587.x","volume":"69","author":"T Gneiting","year":"2007","unstructured":"Gneiting, T., Balabdaoui, F., Raftery, A.E.: Probabilistic forecasts, calibration and sharpness. J. R. Stat. Soc. Ser. B Stat Methodol. 69(2), 243\u2013268 (2007)","journal-title":"J. R. Stat. Soc. Ser. B Stat Methodol."},{"key":"10895_CR21","unstructured":"Hampel , F.R., Ronchetti, E.M., Rousseeuw, P.J., et\u00a0al.: Robust Statistics: The Approach Based on Influence Functions. John Wiley & Sons (1986)"},{"key":"10895_CR22","doi-asserted-by":"crossref","unstructured":"Hamura, Y., Irie, K., Sugasawa, S.: Robust bayesian modeling of counts with zero inflation and outliers: Theoretical robustness and efficient computation. Journal of the American Statistical Association pp 1\u201313 (2025)","DOI":"10.1080\/01621459.2024.2447111"},{"key":"10895_CR23","doi-asserted-by":"publisher","DOI":"10.7717\/peerj.616","volume":"2","author":"XA Harrison","year":"2014","unstructured":"Harrison, X.A.: Using observation-level random effects to model overdispersion in count data in ecology and evolution. PeerJ 2, e616 (2014)","journal-title":"PeerJ"},{"issue":"4","key":"10895_CR24","doi-asserted-by":"publisher","first-page":"753","DOI":"10.1007\/s11222-018-9836-0","volume":"29","author":"M Hartmann","year":"2019","unstructured":"Hartmann, M., Vanhatalo, J.: Laplace approximation and natural gradient for gaussian process regression with heteroscedastic student-t model. Stat. Comput. 29(4), 753\u2013773 (2019)","journal-title":"Stat. Comput."},{"key":"10895_CR25","unstructured":"Hawryluk, I., Hoeltgebaum, H., Mishra, S., et\u00a0al.: Gaussian process nowcasting: application to covid-19 mortality reporting. In: Uncertainty in Artificial Intelligence, PMLR, pp 1258\u20131268 (2021)"},{"key":"10895_CR26","unstructured":"Hensman, J., Matthews, A., Ghahramani, Z.: Scalable variational gaussian process classification. In: Artificial Intelligence and Statistics, PMLR, pp 351\u2013360 (2015)"},{"key":"10895_CR27","unstructured":"Jyl\u00e4nki, P., Vanhatalo, J., Vehtari, A.: Robust gaussian process regression with a student-t likelihood. Journal of Machine Learning Research 12(11) (2011)"},{"issue":"13","key":"10895_CR28","doi-asserted-by":"publisher","first-page":"2403","DOI":"10.1002\/sim.9361","volume":"41","author":"EC Kenne Pagui","year":"2022","unstructured":"Kenne Pagui, E.C., Salvan, A., Sartori, N.: Improved estimation in negative binomial regression. Stat. Med. 41(13), 2403\u20132416 (2022)","journal-title":"Stat. Med."},{"key":"10895_CR29","unstructured":"Knoblauch, J.: Robust deep gaussian processes. arXiv preprint arXiv:1904.02303 (2019)"},{"issue":"132","key":"10895_CR30","first-page":"1","volume":"23","author":"J Knoblauch","year":"2022","unstructured":"Knoblauch, J., Jewson, J., Damoulas, T.: An optimization-centric view on bayes\u2019 rule: Reviewing and generalizing variational inference. J. Mach. Learn. Res. 23(132), 1\u2013109 (2022)","journal-title":"J. Mach. Learn. Res."},{"key":"10895_CR31","unstructured":"Law, H.C., Sejdinovic, D., Cameron, E., et\u00a0al.: Variational learning on aggregate outputs with gaussian processes. Advances in neural information processing systems 31 (2018)"},{"issue":"22","key":"10895_CR32","doi-asserted-by":"publisher","first-page":"4129","DOI":"10.1093\/bioinformatics\/btab455","volume":"37","author":"Q Li","year":"2021","unstructured":"Li, Q., Zhang, M., Xie, Y., et al.: Bayesian modeling of spatial molecular profiling data via gaussian process. Bioinformatics 37(22), 4129\u20134136 (2021)","journal-title":"Bioinformatics"},{"issue":"2","key":"10895_CR33","doi-asserted-by":"publisher","first-page":"503","DOI":"10.1016\/S0378-3758(02)00265-3","volume":"116","author":"CH M\u00fcller","year":"2003","unstructured":"M\u00fcller, C.H., Neykov, N.: Breakdown points of trimmed likelihood estimators and related estimators in generalized linear models. Journal of Statistical Planning and inference 116(2), 503\u2013519 (2003)","journal-title":"Journal of Statistical Planning and inference"},{"key":"10895_CR34","unstructured":"Murphy, K.P.: Probabilistic machine learning: Advanced topics. MIT press (2023)"},{"key":"10895_CR35","first-page":"1065","volume":"20","author":"A Naish-Guzman","year":"2007","unstructured":"Naish-Guzman, A., Holden, S.: Robust regression with twinned gaussian processes. Adv. Neural. Inf. Process. Syst. 20, 1065\u20131072 (2007)","journal-title":"Adv. Neural. Inf. Process. Syst."},{"key":"10895_CR36","unstructured":"Nickisch, H., Solin, A., Grigorevskiy, A.: State space gaussian processes with non-gaussian likelihood. In: International Conference on Machine Learning, PMLR, pp 3789\u20133798 (2018)"},{"key":"10895_CR37","volume-title":"Gaussian processes for machine learning","author":"CE Rasmussen","year":"2006","unstructured":"Rasmussen, C.E., Williams, C.K.: Gaussian processes for machine learning. MIT Press (2006)"},{"key":"10895_CR38","unstructured":"Wang, Y., Kucukelbir, A., Blei, D.M.: Robust probabilistic modeling with bayesian data reweighting. In: International Conference on Machine Learning, PMLR, pp 3646\u20133655 (2017)"},{"key":"10895_CR39","unstructured":"Xiong, L., Zhu, F.: Minimum density power divergence estimator for negative binomial integer-valued garch models. Communications in Mathematics and Statistics pp 1\u201329 (2022)"},{"key":"10895_CR40","doi-asserted-by":"crossref","unstructured":"Zandkarimi, E., Moghimbeigi, A., Mahjub, H., et\u00a0al.: Robust inference in the multilevel zero-inflated negative binomial model. Journal of Applied Statistics (2020)","DOI":"10.1080\/02664763.2019.1636942"},{"key":"10895_CR41","doi-asserted-by":"publisher","first-page":"1","DOI":"10.18637\/jss.v027.i08","volume":"27","author":"A Zeileis","year":"2008","unstructured":"Zeileis, A., Kleiber, C., Jackman, S.: Regression models for count data in r. J. Stat. Softw. 27, 1\u201325 (2008)","journal-title":"J. Stat. Softw."},{"issue":"3","key":"10895_CR42","doi-asserted-by":"publisher","first-page":"883","DOI":"10.1080\/10618600.2023.2289532","volume":"33","author":"Z Zhang","year":"2024","unstructured":"Zhang, Z., Stringer, A., Brown, P., et al.: Model-based smoothing with integrated wiener processes and overlapping splines. J. Comput. Graph. Stat. 33(3), 883\u2013895 (2024)","journal-title":"J. Comput. Graph. Stat."},{"key":"10895_CR43","doi-asserted-by":"crossref","unstructured":"Zhang, Z., Brown, P., Stafford, J.: Efficient modeling of quasi-periodic data with seasonal gaussian process. Stat. Comput. 35(2), 32 (2025)","DOI":"10.1007\/s11222-025-10565-2"}],"container-title":["Statistics and Computing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11222-026-10895-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s11222-026-10895-9","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11222-026-10895-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,14]],"date-time":"2026-05-14T04:26:02Z","timestamp":1778732762000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s11222-026-10895-9"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,14]]},"references-count":43,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2026,8]]}},"alternative-id":["10895"],"URL":"https:\/\/doi.org\/10.1007\/s11222-026-10895-9","relation":{},"ISSN":["0960-3174","1573-1375"],"issn-type":[{"value":"0960-3174","type":"print"},{"value":"1573-1375","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,14]]},"assertion":[{"value":"12 July 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"26 April 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"14 May 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The author has no relevant financial or non-financial interests to disclose.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}},{"value":"The authors declare no competing interests.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"139"}}