{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,27]],"date-time":"2026-09-27T01:12:00Z","timestamp":1790471520566,"version":"4.1.0"},"reference-count":59,"publisher":"Society for Industrial & Applied Mathematics (SIAM)","issue":"4","funder":[{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["DMS-1712822"],"award-info":[{"award-number":["DMS-1712822"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["IIS-1632971"],"award-info":[{"award-number":["IIS-1632971"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["SIAM J. Optim."],"published-print":{"date-parts":[[2018,1]]},"abstract":"<jats:p>Many modern statistical estimation problems are defined by three major components: a statistical model that postulates the dependence of an output variable on the input features; a loss function measuring the error between the observed output and the model predicted output; and a regularizer that controls the overfitting and\/or variable selection in the model. We study the sampling version of this generic statistical estimation problem where the model parameters are estimated by empirical risk minimization, which involves the minimization of the empirical average of the loss function at the data points weighted by the model regularizer. In our setup we allow all three component functions discussed above to be of the difference-of-convex (dc) type and illustrate them with a host of commonly used examples, including those in continuous piecewise affine regression and in deep learning (where the activation functions are piecewise affine). We describe a nonmonotone majorization-minimization (MM) algorithm for solving the unified nonconvex, nondifferentiable optimization problem which is formulated as a specially structured composite dc program of the pointwise max type, and present convergence results to a directional stationary solution. An efficient semismooth Newton method is proposed to solve the dual of the MM subproblems. Numerical results are presented to demonstrate the effectiveness of the proposed algorithm and the superiority of continuous piecewise affine regression over the standard linear model.<\/jats:p>","DOI":"10.1137\/18m117337x","type":"journal-article","created":{"date-parts":[[2018,12,13]],"date-time":"2018-12-13T10:52:51Z","timestamp":1544698371000},"page":"3344-3374","source":"Crossref","is-referenced-by-count":52,"title":["Composite Difference-Max Programs for Modern Statistical Estimation Problems"],"prefix":"10.1137","volume":"28","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4173-5647","authenticated-orcid":true,"given":"Ying","family":"Cui","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jong-Shi","family":"Pang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bodhisattva","family":"Sen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"351","published-online":{"date-parts":[[2018,12,13]]},"reference":[{"key":"atypb1","doi-asserted-by":"publisher","DOI":"10.1137\/16M1084754"},{"key":"atypb2","doi-asserted-by":"publisher","DOI":"10.1007\/s10107-007-0133-5"},{"key":"atypb3","doi-asserted-by":"publisher","DOI":"10.1007\/s10107-011-0484-9"},{"key":"atypb4","doi-asserted-by":"publisher","DOI":"10.1007\/s10589-008-9174-9"},{"key":"atypb5","doi-asserted-by":"crossref","unstructured":"H.H. Bauschke and P.L. Combettes,\n                      Convex Analysis and Monotone Operator Theory in Hilbert Spaces\n                      , LMS Books Math., Springer, New York, 2011.","DOI":"10.1007\/978-1-4419-9467-7"},{"key":"atypb6","unstructured":"D.P. Bertsekas,\n                      Nonlinear Programming\n                      , 3rd ed., Athena Scientific, Belmont, MA, 2016."},{"key":"atypb7","unstructured":"P.J. Bickel and K. Doksum,\n                      Mathematical Statistics, Basic Ideas and Selected Topics\n                      , Prentice Hall, Upper Saddle River, NJ, 2006."},{"key":"atypb8","doi-asserted-by":"publisher","DOI":"10.1137\/050644641"},{"key":"atypb9","doi-asserted-by":"publisher","DOI":"10.1287\/moor.2015.0735"},{"key":"atypb10","doi-asserted-by":"publisher","DOI":"10.1007\/s10107-013-0701-9"},{"key":"atypb11","unstructured":"L.D. Brown,\n                      Fundamentals of Statistical Exponential Families\n                      , IMS Lect. Notes Monogr. Ser., Institute of Mathematical Statistics, Hayward, CA, 1986."},{"key":"atypb12","doi-asserted-by":"publisher","DOI":"10.1007\/s00041-008-9045-x"},{"key":"atypb13","doi-asserted-by":"crossref","unstructured":"F.H. Clarke,\n                      Optimization and Nonsmooth Analysis\n                      , Classics Appl. Math. 5, SIAM, Philadelphia, 1990.","DOI":"10.1137\/1.9781611971309"},{"key":"atypb14","doi-asserted-by":"publisher","DOI":"10.1023\/A:1022627411411"},{"key":"atypb15","unstructured":"Y. Cui and J.S. Pang,\n                      On the Finite Number of Directional Stationary Values of Piecewise Programs\n                      ,https:\/\/arxiv.org\/abs\/1803.00190(2018)."},{"key":"atypb16","unstructured":"Y. Cui, T.H. Chang, M. Hong, and J.S. Pang,\n                      A Study of Piecewise Linear-Quadratic Programs\n                      ,https:\/\/arxiv.org\/abs\/1709.05758(2018)."},{"key":"atypb17","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1111\/j.2517-6161.1977.tb01600.x","volume":"39","author":"Dempster A.P.","year":"1977","journal-title":"J. Roy. Statist. Soc. B"},{"key":"atypb19","unstructured":"D. Dunson and L.A. Hannah,\n                      Bayesian Nonparametric Multivariate Convex Regression\n                      ,https:\/\/arxiv.org\/abs\/1109.0322(2011)."},{"key":"atypb20","first-page":"3261","volume":"14","author":"Dunson D.","year":"2013","journal-title":"J. Mach. Learn. Res."},{"key":"atypb21","doi-asserted-by":"crossref","unstructured":"F. Facchinei and J.S. Pang,\n                      Finite-Dimensional Variational Inequalities and Complementarity Problems\n                      , Springer, New York, 2003.","DOI":"10.1007\/b97543"},{"key":"atypb22","doi-asserted-by":"publisher","DOI":"10.1198\/016214501753382273"},{"key":"atypb23","first-page":"315","author":"Glorot X.","year":"2011","journal-title":"MA"},{"key":"atypb24","unstructured":"G. Hahn, M. Banergjee, and B. Sen,\n                      Parameter estimation and inference in a continuous piecewise linear regression model\n                      , Technical report, Columbia University, New York, 2017."},{"key":"atypb25","first-page":"707","volume":"9","author":"Hartman P.","year":"1959","journal-title":"Pac. J. Optim."},{"key":"atypb26","doi-asserted-by":"crossref","unstructured":"T. Hastie, R. Tibshirani, and M. Weinwright,\n                      Statistical Learning with Sparsity: The Lasso and Generalizations\n                      , CRC Press, Boca Raton, FL, 2015.","DOI":"10.1201\/b18401"},{"key":"atypb27","doi-asserted-by":"publisher","DOI":"10.1198\/0003130042836"},{"key":"atypb28","doi-asserted-by":"publisher","DOI":"10.2307\/1913643"},{"key":"atypb29","first-page":"352","volume":"29","author":"Kojima M.","year":"1986","journal-title":"J. Oper. Res. Soc. Japan"},{"key":"atypb30","first-page":"114","author":"Kummer B.","year":"1988","journal-title":"Berlin"},{"key":"atypb31","doi-asserted-by":"crossref","unstructured":"K. Lange,\n                      MM Optimization Algorithms\n                      , Other Titles Appl. Math. 147, SIAM, Philadelphia, 2016.","DOI":"10.1137\/1.9781611974409"},{"key":"atypb32","doi-asserted-by":"publisher","DOI":"10.1007\/s10957-018-1345-y"},{"key":"atypb33","first-page":"25","volume":"133","author":"Le Thi H.A.","year":"2005","journal-title":"Ann. Oper. Res."},{"key":"atypb34","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejor.2014.11.031"},{"key":"atypb35","doi-asserted-by":"publisher","DOI":"10.1137\/16M1097572"},{"key":"atypb37","first-page":"783","author":"Mairal J.","year":"2013","journal-title":"NY"},{"key":"atypb38","doi-asserted-by":"publisher","DOI":"10.1137\/140957639"},{"key":"atypb40","doi-asserted-by":"publisher","DOI":"10.1137\/0315061"},{"key":"atypb41","doi-asserted-by":"publisher","DOI":"10.1137\/120892167"},{"key":"atypb42","doi-asserted-by":"crossref","unstructured":"B.S. Mordukhovich,\n                      Variational Analysis and Generalized Differentiation\n                      I, Springer, New York, 2006.","DOI":"10.1007\/3-540-31247-1"},{"key":"atypb43","first-page":"807","author":"Nair V.","year":"2010","journal-title":"WI"},{"key":"atypb44","doi-asserted-by":"crossref","unstructured":"J.M. Ortega and W.C. Rheinboldt,\n                      Iterative Solution of Nonlinear Equations in Several Variables\n                      , Classics Appl. Math. 30, SIAM, Philadelphia, 2000.","DOI":"10.1137\/1.9780898719468"},{"key":"atypb45","doi-asserted-by":"publisher","DOI":"10.1287\/moor.15.2.311"},{"key":"atypb46","doi-asserted-by":"publisher","DOI":"10.1007\/BF02193060"},{"key":"atypb47","doi-asserted-by":"publisher","DOI":"10.1287\/moor.2016.0795"},{"key":"atypb48","doi-asserted-by":"publisher","DOI":"10.1137\/050624509"},{"key":"atypb49","doi-asserted-by":"publisher","DOI":"10.1007\/BF01581275"},{"key":"atypb50","doi-asserted-by":"crossref","unstructured":"R.T. Rockafellar and R.J.B. Wets,\n                      Variational Analysis\n                      , Springer, New York, 1998.","DOI":"10.1007\/978-3-642-02431-3"},{"key":"atypb51","unstructured":"S. Scholtes,\n                      Introduction to Piecewise Differentiable Equations\n                      , SpringerBriefs Optim., Springer, New York, 2002."},{"key":"atypb52","unstructured":"J. Sun,\n                      On Monotropic Piecewise Quadratic Programming\n                      , Ph.D. dissertation, Department of Applied Mathematics, University of Washington, Seattle, WA, 1986."},{"key":"atypb53","doi-asserted-by":"publisher","DOI":"10.1007\/BF00939839"},{"key":"atypb54","doi-asserted-by":"publisher","DOI":"10.1111\/j.1467-9868.2005.00490.x"},{"key":"atypb55","doi-asserted-by":"publisher","DOI":"10.1007\/s10589-017-9954-1"},{"key":"atypb56","doi-asserted-by":"publisher","DOI":"10.1214\/aos\/1176346060"},{"key":"atypb57","doi-asserted-by":"publisher","DOI":"10.1007\/s12532-015-0082-6"},{"key":"atypb58","doi-asserted-by":"publisher","DOI":"10.1137\/140952363"},{"key":"atypb59","doi-asserted-by":"crossref","unstructured":"D. Yu and L. Deng,\n                      Automatic Speech Recognition: A Deep Learning Approach\n                      , Signals Commun. Technol., Springer, London, 2015.","DOI":"10.1007\/978-1-4471-5779-3"},{"key":"atypb60","doi-asserted-by":"publisher","DOI":"10.1214\/09-AOS729"},{"key":"atypb61","doi-asserted-by":"publisher","DOI":"10.1007\/s10107-017-1209-5"},{"key":"atypb62","doi-asserted-by":"publisher","DOI":"10.1137\/080718206"}],"container-title":["SIAM Journal on Optimization"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/epubs.siam.org\/doi\/pdf\/10.1137\/18M117337X","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T18:52:58Z","timestamp":1787338378000},"score":1,"resource":{"primary":{"URL":"https:\/\/epubs.siam.org\/doi\/10.1137\/18M117337X"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2018,1]]},"references-count":59,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2018,1]]}},"alternative-id":["10.1137\/18M117337X"],"URL":"https:\/\/doi.org\/10.1137\/18m117337x","relation":{},"ISSN":["1052-6234","1095-7189"],"issn-type":[{"value":"1052-6234","type":"print"},{"value":"1095-7189","type":"electronic"}],"subject":[],"published":{"date-parts":[[2018,1]]}}}