{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,4]],"date-time":"2026-06-04T17:02:10Z","timestamp":1780592530736,"version":"3.54.1"},"reference-count":35,"publisher":"Springer Science and Business Media LLC","issue":"2","license":[{"start":{"date-parts":[[2025,11,26]],"date-time":"2025-11-26T00:00:00Z","timestamp":1764115200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,11,26]],"date-time":"2025-11-26T00:00:00Z","timestamp":1764115200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100004434","name":"Universit\u00e0 degli Studi di Firenze","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100004434","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Comput Optim Appl"],"published-print":{"date-parts":[[2026,3]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>In this work, we consider smooth unconstrained optimization problems and we deal with the class of gradient methods with momentum, i.e., descent algorithms where the search direction is defined as a linear combination of the current gradient and the preceding search direction. This family of algorithms includes nonlinear conjugate gradient methods and Polyak\u2019s heavy-ball approach, and is thus of high practical and theoretical interest in large-scale nonlinear optimization. We propose a general framework where the scalars of the linear combination defining the search direction are computed simultaneously by minimizing the approximate quadratic model in the 2 dimensional subspace. This strategy allows us to define a class of gradient methods with momentum enjoying global convergence guarantees and an optimal worst-case complexity bound in the nonconvex setting. Differently than all related works in the literature, the convergence conditions are stated in terms of the Hessian matrix of the bi-dimensional quadratic model. To the best of our knowledge, these results are novel to the literature. Moreover, extensive computational experiments show that the gradient method with momentum here presented is competitive with respect to other popular solvers for nonconvex unconstrained problems.<\/jats:p>","DOI":"10.1007\/s10589-025-00741-5","type":"journal-article","created":{"date-parts":[[2025,11,26]],"date-time":"2025-11-26T13:18:53Z","timestamp":1764163133000},"page":"795-820","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":2,"title":["A globally convergent gradient method with momentum"],"prefix":"10.1007","volume":"93","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2488-5486","authenticated-orcid":false,"given":"Matteo","family":"Lapucci","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Giampaolo","family":"Liuzzi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Stefano","family":"Lucidi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Davide","family":"Pucci","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marco","family":"Sciandrone","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,11,26]]},"reference":[{"issue":"2","key":"741_CR1","doi-asserted-by":"publisher","first-page":"223","DOI":"10.1137\/16M1080173","volume":"60","author":"L Bottou","year":"2018","unstructured":"Bottou, L., Curtis, F.E., Nocedal, J.: Optimization methods for large-scale machine learning. SIAM Rev. 60(2), 223\u2013311 (2018)","journal-title":"SIAM Rev."},{"key":"741_CR2","doi-asserted-by":"publisher","DOI":"10.1017\/9781009004282","volume-title":"Optimization for Data Analysis","author":"SJ Wright","year":"2022","unstructured":"Wright, S.J., Recht, B.: Optimization for Data Analysis. Cambridge University Press, Cambridge (2022)"},{"issue":"5","key":"741_CR3","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/0041-5553(64)90137-5","volume":"4","author":"BT Polyak","year":"1964","unstructured":"Polyak, B.T.: Some methods of speeding up the convergence of iteration methods. USSR Comput. Math. Math. Phys. 4(5), 1\u201317 (1964)","journal-title":"USSR Comput. Math. Math. Phys."},{"key":"741_CR4","unstructured":"Polyak, B.T.: Introduction to optimization. Optimization software. Inc., Publications Division, New York 1, 32 (1987)"},{"key":"741_CR5","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-26790-1","volume-title":"Introduction to Methods for Nonlinear Optimization","author":"L Grippo","year":"2023","unstructured":"Grippo, L., Sciandrone, M.: Introduction to Methods for Nonlinear Optimization. Springer, Cham (2023)"},{"key":"741_CR6","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejco.2022.100044","volume":"10","author":"R Chan-Renous-Legoubin","year":"2022","unstructured":"Chan-Renous-Legoubin, R., Royer, C.W.: A nonlinear conjugate gradient method with complexity guarantees and its application to nonconvex regression. EURO Journal on Computational Optimization 10, 100044 (2022)","journal-title":"EURO Journal on Computational Optimization"},{"key":"741_CR7","doi-asserted-by":"crossref","unstructured":"Neumaier, A., Kimiaei, M., Azmi, B.: Globally linearly convergent nonlinear conjugate gradients without wolfe line search. Numer. Algor., 1\u201327 (2024)","DOI":"10.1007\/s11075-024-01764-5"},{"issue":"2","key":"741_CR8","doi-asserted-by":"publisher","first-page":"820","DOI":"10.1007\/s10957-023-02325-x","volume":"200","author":"Z Liu","year":"2024","unstructured":"Liu, Z., Ni, Y., Liu, H., Sun, W.: A new subspace minimization conjugate gradient method for unconstrained minimization. J. Optim. Theory Appl. 200(2), 820\u2013851 (2024)","journal-title":"J. Optim. Theory Appl."},{"key":"741_CR9","doi-asserted-by":"publisher","first-page":"811","DOI":"10.1007\/s10957-021-01897-w","volume":"190","author":"W Sun","year":"2021","unstructured":"Sun, W., Liu, H., Liu, Z.: A class of accelerated subspace minimization conjugate gradient methods. J. Optim. Theory Appl. 190, 811\u2013840 (2021)","journal-title":"J. Optim. Theory Appl."},{"key":"741_CR10","doi-asserted-by":"publisher","first-page":"813","DOI":"10.1007\/s11075-017-0284-2","volume":"76","author":"Y Yang","year":"2017","unstructured":"Yang, Y., Chen, Y., Lu, Y.: A subspace conjugate gradient algorithm for large-scale unconstrained optimization. Numer. Algor. 76, 813\u2013828 (2017)","journal-title":"Numer. Algor."},{"issue":"1","key":"741_CR11","first-page":"69","volume":"75","author":"Y-X Yuan","year":"1995","unstructured":"Yuan, Y.-X., Stoer, J.: A subspace study on conjugate gradient algorithms. ZAMM-J. Appl. Math. Mech.\/Zeitschrift f\u00fcr Angewandte Mathematik und Mechanik 75(1), 69\u201377 (1995)","journal-title":"ZAMM-J. Appl. Math. Mech.\/Zeitschrift f\u00fcr Angewandte Mathematik und Mechanik"},{"key":"741_CR12","volume-title":"Optimization for Machine Learning","author":"S Sra","year":"2012","unstructured":"Sra, S., Nowozin, S., Wright, S.J.: Optimization for Machine Learning. Mit Press, Cambridge (2012)"},{"key":"741_CR13","doi-asserted-by":"crossref","unstructured":"Ghadimi, E., Feyzmahdavian, H.R., Johansson, M.: Global convergence of the heavy-ball method for convex optimization. In: 2015 European Control Conference (ECC), pp. 310\u2013315 (2015). IEEE","DOI":"10.1109\/ECC.2015.7330562"},{"issue":"1","key":"741_CR14","doi-asserted-by":"publisher","first-page":"57","DOI":"10.1137\/15M1009597","volume":"26","author":"L Lessard","year":"2016","unstructured":"Lessard, L., Recht, B., Packard, A.: Analysis and design of optimization algorithms via integral quadratic constraints. SIAM J. Optim. 26(1), 57\u201395 (2016)","journal-title":"SIAM J. Optim."},{"issue":"9","key":"741_CR15","doi-asserted-by":"publisher","first-page":"3245","DOI":"10.1007\/s10994-022-06215-7","volume":"111","author":"S Saab Jr","year":"2022","unstructured":"Saab, S., Jr., Phoha, S., Zhu, M., Ray, A.: An adaptive Polyak heavy-ball method. Mach. Learn. 111(9), 3245\u20133277 (2022)","journal-title":"Mach. Learn."},{"key":"741_CR16","first-page":"18261","volume":"33","author":"Y Liu","year":"2020","unstructured":"Liu, Y., Gao, Y., Yin, W.: An improved analysis of stochastic gradient descent with momentum. Adv. Neural. Inf. Process. Syst. 33, 18261\u201318271 (2020)","journal-title":"Adv. Neural. Inf. Process. Syst."},{"key":"741_CR17","unstructured":"Sutskever, I., Martens, J., Dahl, G., Hinton, G.: On the importance of initialization and momentum in deep learning. In: International Conference on Machine Learning, pp. 1139\u20131147 (2013). PMLR"},{"issue":"1\u20132","key":"741_CR18","doi-asserted-by":"publisher","first-page":"71","DOI":"10.1007\/s10107-019-01406-y","volume":"184","author":"Y Carmon","year":"2020","unstructured":"Carmon, Y., Duchi, J.C., Hinder, O., Sidford, A.: Lower bounds for finding stationary points I. Math. Program. 184(1\u20132), 71\u2013120 (2020)","journal-title":"Math. Program."},{"issue":"1\u20133","key":"741_CR19","doi-asserted-by":"publisher","first-page":"503","DOI":"10.1007\/BF01589116","volume":"45","author":"DC Liu","year":"1989","unstructured":"Liu, D.C., Nocedal, J.: On the limited memory BFGS method for large scale optimization. Math. Program. 45(1\u20133), 503\u2013528 (1989)","journal-title":"Math. Program."},{"issue":"1","key":"741_CR20","doi-asserted-by":"publisher","first-page":"113","DOI":"10.1145\/1132973.1132979","volume":"32","author":"WW Hager","year":"2006","unstructured":"Hager, W.W., Zhang, H.: Algorithm 851: Cg_descent, a conjugate gradient method with guaranteed descent. ACM Trans. Math. Softw. (TOMS) 32(1), 113\u2013137 (2006)","journal-title":"ACM Trans. Math. Softw. (TOMS)"},{"issue":"4","key":"741_CR21","doi-asserted-by":"publisher","first-page":"2150","DOI":"10.1137\/120898097","volume":"23","author":"WW Hager","year":"2013","unstructured":"Hager, W.W., Zhang, H.: The limited memory conjugate gradient method. SIAM J. Optim. 23(4), 2150\u20132168 (2013)","journal-title":"SIAM J. Optim."},{"issue":"3","key":"741_CR22","doi-asserted-by":"publisher","first-page":"543","DOI":"10.1007\/s12532-022-00219-z","volume":"14","author":"C-P Lee","year":"2022","unstructured":"Lee, C.-P., Wang, P.-W., Lin, C.-J.: Limited-memory common-directions method for large-scale optimization: convergence, parallelization, and distributed optimization. Math. Program. Comput. 14(3), 543\u2013591 (2022)","journal-title":"Math. Program. Comput."},{"key":"741_CR23","unstructured":"Zhang, C., Ge, D., He, C., Jiang, B., Jiang, Y., Ye, Y.: Drsom: A dimension reduced second-order method. arXiv preprint arXiv:2208.00208 (2022)"},{"issue":"3","key":"741_CR24","doi-asserted-by":"publisher","first-page":"2025","DOI":"10.1137\/23M1567229","volume":"46","author":"T Tang","year":"2024","unstructured":"Tang, T., Toh, K.-C., Xiao, N., Ye, Y.: A Riemannian dimension-reduced second-order method with application in sensor network localization. SIAM J. Sci. Comput. 46(3), 2025\u20132046 (2024)","journal-title":"SIAM J. Sci. Comput."},{"key":"741_CR25","doi-asserted-by":"publisher","DOI":"10.1137\/1.9781611976991","volume-title":"Evaluation Complexity of Algorithms for Nonconvex Optimization: Theory, Computation and Perspectives","author":"C Cartis","year":"2022","unstructured":"Cartis, C., Gould, N.I.M., Toint, P.L.: Evaluation Complexity of Algorithms for Nonconvex Optimization: Theory, Computation and Perspectives. SIAM, Philadelphia (2022)"},{"issue":"5","key":"741_CR26","doi-asserted-by":"publisher","first-page":"1349","DOI":"10.1080\/02331934.2013.869809","volume":"64","author":"C Cartis","year":"2015","unstructured":"Cartis, C., Sampaio, P.R., Toint, P.L.: Worst-case evaluation complexity of non-monotone gradient-related algorithms for unconstrained optimization. Optimization 64(5), 1349\u20131361 (2015)","journal-title":"Optimization"},{"key":"741_CR27","unstructured":"Bertsekas, D.P.: Nonlinear Programming, Third Edition vol. 4, 2nd edn. Athena Scientific, Belmont, Massachusetts (2016)"},{"key":"741_CR28","doi-asserted-by":"publisher","first-page":"401","DOI":"10.1007\/BF00940345","volume":"60","author":"L Grippo","year":"1989","unstructured":"Grippo, L., Lampariello, F., Lucidi, S.: A truncated Newton method with nonmonotone line search for unconstrained optimization. J. Optim. Theory Appl. 60, 401\u2013419 (1989)","journal-title":"J. Optim. Theory Appl."},{"issue":"2","key":"741_CR29","doi-asserted-by":"publisher","first-page":"400","DOI":"10.1137\/0719025","volume":"19","author":"RS Dembo","year":"1982","unstructured":"Dembo, R.S., Eisenstat, S.C., Steihaug, T.: Inexact Newton methods. SIAM J. Numer. Anal. 19(2), 400\u2013408 (1982)","journal-title":"SIAM J. Numer. Anal."},{"issue":"1","key":"741_CR30","doi-asserted-by":"publisher","first-page":"26","DOI":"10.1137\/S1052623494266365","volume":"7","author":"M Raydan","year":"1997","unstructured":"Raydan, M.: The Barzilai and Borwein gradient method for the large scale unconstrained minimization problem. SIAM J. Optim. 7(1), 26\u201333 (1997)","journal-title":"SIAM J. Optim."},{"issue":"4","key":"741_CR31","doi-asserted-by":"publisher","first-page":"1043","DOI":"10.1137\/S1052623403428208","volume":"14","author":"H Zhang","year":"2004","unstructured":"Zhang, H., Hager, W.W.: A nonmonotone line search technique and its application to unconstrained optimization. SIAM J. Optim. 14(4), 1043\u20131056 (2004)","journal-title":"SIAM J. Optim."},{"issue":"1","key":"741_CR32","first-page":"35","volume":"2","author":"WW Hager","year":"2006","unstructured":"Hager, W.W., Zhang, H.: A survey of nonlinear conjugate gradient methods. Pac. J. Optim. 2(1), 35\u201358 (2006)","journal-title":"Pac. J. Optim."},{"issue":"1","key":"741_CR33","doi-asserted-by":"publisher","first-page":"170","DOI":"10.1137\/030601880","volume":"16","author":"WW Hager","year":"2005","unstructured":"Hager, W.W., Zhang, H.: A new conjugate gradient method with guaranteed descent and an efficient line search. SIAM J. Optim. 16(1), 170\u2013192 (2005)","journal-title":"SIAM J. Optim."},{"key":"741_CR34","doi-asserted-by":"publisher","first-page":"201","DOI":"10.1007\/s101070100263","volume":"91","author":"ED Dolan","year":"2002","unstructured":"Dolan, E.D., Mor\u00e9, J.J.: Benchmarking optimization software with performance profiles. Math. Program. 91, 201\u2013213 (2002)","journal-title":"Math. Program."},{"key":"741_CR35","unstructured":"Gould, N.I.M., Orban, D., Toint, Ph.L.: The Constrained and Unconstrained Testing Environment with safe threads (CUTEst) for optimization software. https:\/\/github.com\/ralna\/CUTEst (2019)"}],"container-title":["Computational Optimization and Applications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10589-025-00741-5.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s10589-025-00741-5","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10589-025-00741-5.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,29]],"date-time":"2026-01-29T12:01:48Z","timestamp":1769688108000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s10589-025-00741-5"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,26]]},"references-count":35,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2026,3]]}},"alternative-id":["741"],"URL":"https:\/\/doi.org\/10.1007\/s10589-025-00741-5","relation":{},"ISSN":["0926-6003","1573-2894"],"issn-type":[{"value":"0926-6003","type":"print"},{"value":"1573-2894","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,11,26]]},"assertion":[{"value":"29 September 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"3 October 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"26 November 2025","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 authors have no Conflict of interest to declare that are relevant to the content of this article.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}}]}}