{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T14:33:50Z","timestamp":1787322830729,"version":"build-2736575974"},"reference-count":57,"publisher":"Society for Industrial & Applied Mathematics (SIAM)","issue":"1","funder":[{"DOI":"10.13039\/501100001711","name":"Swiss National Science Foundation","doi-asserted-by":"crossref","award":["200020 184646\/1"],"award-info":[{"award-number":["200020 184646\/1"]}],"id":[{"id":"10.13039\/501100001711","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100010663","name":"H2020 European Research Council","doi-asserted-by":"publisher","award":["101020573"],"award-info":[{"award-number":["101020573"]}],"id":[{"id":"10.13039\/100010663","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["SIAM J. Imaging Sci."],"published-print":{"date-parts":[[2024,3,31]]},"abstract":"<jats:p>Abstract.<\/jats:p>\n                  <jats:p>We propose to learn non-convex regularizers with a prescribed upper bound on their weak-convexity modulus. Such regularizers give rise to variational denoisers that minimize a convex energy. They rely on few parameters (less than 15,000) and offer a signal-processing interpretation as they mimic handcrafted sparsity-promoting regularizers. Through numerical experiments, we show that such denoisers outperform convex-regularization methods as well as the popular BM3D denoiser. Additionally, the learned regularizer can be deployed to solve inverse problems with iterative schemes that provably converge. For both CT and MRI reconstruction, the regularizer generalizes well and offers an excellent tradeoff between performance, number of parameters, guarantees, and interpretability when compared to other data-driven approaches.<\/jats:p>","DOI":"10.1137\/23m1565243","type":"journal-article","created":{"date-parts":[[2024,1,18]],"date-time":"2024-01-18T04:02:01Z","timestamp":1705550521000},"page":"91-115","source":"Crossref","is-referenced-by-count":35,"title":["Learning Weakly Convex Regularizers for Convergent Image-Reconstruction Algorithms"],"prefix":"10.1137","volume":"17","author":[{"given":"Alexis","family":"Goujon","sequence":"first","affiliation":[{"name":"Biomedical Imaging Group, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne PFLCH-1015 Lausanne."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sebastian","family":"Neumayer","sequence":"additional","affiliation":[{"name":"Biomedical Imaging Group, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne PFLCH-1015 Lausanne."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michael","family":"Unser","sequence":"additional","affiliation":[{"name":"Biomedical Imaging Group, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne PFLCH-1015 Lausanne."}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"351","published-online":{"date-parts":[[2024,1,18]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6420\/ab551e"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2010.161"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1007\/s10107-011-0484-9"},{"key":"ref4","first-page":"1","volume-title":"Advances in Neural Information Processing Systems","volume":"32","author":"Bai S.","year":"2019"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1137\/080716542"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2007.906002"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/OJSP.2020.3039379"},{"key":"ref8","unstructured":"P. Bohra, D. Perdios, A. Goujon, S. Emery, and M. Unser, Learning Lipschitz-controlled activation functions in neural networks for Plug-and-Play image reconstruction methods, in NeurIPS 2021 Workshop on Deep Learning and Inverse Problems, 2021, pp. 1\u20139."},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/83.862633"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2016.2596743"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2014.2299065"},{"key":"ref12","first-page":"18152","volume-title":"Advances in Neural Information Processing Systems","volume":"34","author":"Cohen R.","year":"2021"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2007.901238"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/18.382009"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2006.871582"},{"key":"ref16","unstructured":"S. Ducotterd, A. Goujon, P. Bohra, D. Perdios, S. Neumayer, and M. Unser, Improving Lipschitz-constrained Neural Networks by Learning Activation Functions, preprint, https:\/\/arxiv.org\/abs\/2210.16222, 2022."},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1007\/s10107-015-0871-8"},{"key":"ref18","unstructured":"A. Goujon, A. Etemadi, and M. Unser, The Role of Depth, Width, and Activation Complexity in the Number of Linear Regions of Neural Networks, https:\/\/arxiv.org\/abs\/2206.08615, 2022."},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/TCI.2023.3306100"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1007\/s10851-020-00951-y"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1007\/s00041-020-09761-7"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-31975-4_29"},{"key":"ref23","unstructured":"S. Hurault, A. Leclaire, and N. Papadakis, Gradient step denoiser for convergent Plug-and-Play, in International Conference on Learning Representations, 2022, pp. 1\u201330, https:\/\/openreview.net\/forum?id=fPhKeld3Okz."},{"key":"ref24","unstructured":"S. Hurault, A. Leclaire, and N. Papadakis, Proximal denoiser for convergent Plug-and-Play optimization with nonconvex regularization, in 39th International Conference on Machine Learning, Proc. Mach. Learn. Res. 162, PMLR, 2022, pp. 9483\u20139505."},{"key":"ref25","unstructured":"D. P. Kingma and J. Ba, Adam: A method for stochastic optimization, in 3rd International Conference on Learning Representations, ICLR, 2015, http:\/\/arxiv.org\/abs\/1412.6980."},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.2020190007"},{"key":"ref27","doi-asserted-by":"crossref","unstructured":"E. Kobler, A. Effland, K. Kunisch, and T. Pock, Total deep variation for linear inverse problems, in Conference on Computer Vision and Pattern Recognition, IEEE, 2020, pp. 7546\u20137555.","DOI":"10.1109\/CVPR42600.2020.00757"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-66709-6_23"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/97.475823"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1137\/18M1199149"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-03009-4_61-1"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1007\/s10107-019-01374-3"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6420\/ab6d57"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2021.3124600"},{"key":"ref35","first-page":"8516","volume-title":"Advances in Neural Information Processing Systems","volume":"31","author":"Lunz S.","year":"2018"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1561\/2000000101"},{"key":"ref37","first-page":"3759","volume":"43","author":"McCollough C.","year":"2016","journal-title":"Med. Phys."},{"key":"ref38","unstructured":"S. Mukherjee, S. Dittmer, Z. Shumaylov, S. Lunz, O. \u00d6ktem, and C.B. Sch\u00f6nlieb, Learned Convex Regularizers for Inverse Problems, arXiv:2008.02839, 2021."},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2022.3207451"},{"key":"ref40","first-page":"1","volume-title":"NeurIPS Workshop on Deep Learning and Inverse Problems","author":"Mukherjee S.","year":"2021"},{"key":"ref41","first-page":"543","volume":"269","author":"Nesterov Y. E.","year":"1983","journal-title":"Dokl. Akad. Nauk"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1137\/22M1504573"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4939-0790-8_5"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1007\/s10957-018-1272-y"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1137\/130942954"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1007\/s10208-013-9150-3"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/LSP.2015.2432095"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2008.923099"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-008-0197-6"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1016\/0167-2789(92)90242-F"},{"key":"ref51","doi-asserted-by":"crossref","unstructured":"G. Scrivanti, E. Chouzenoux, and J.C. Pesquet, A CNC approach for directional total variation, in 30th European Signal Processing Conference, IEEE, 2022, pp. 488\u2013492, https:\/\/doi.org\/10.23919\/EUSIPCO55093.2022.9909763.","DOI":"10.23919\/EUSIPCO55093.2022.9909763"},{"key":"ref52","unstructured":"H. Sedghi, V. Gupta, and P. M. Long, The singular values of convolutional layers, in International Conference on Learning Representations, 2019, pp. 1\u201312, https:\/\/openreview.net\/forum?id=rJevYoA9Fm."},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2020.07.085"},{"key":"ref54","first-page":"1035","volume":"4","author":"Tikhonov A. N.","year":"1963","journal-title":"Soviet Math."},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3088914"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/TCI.2016.2644865"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2018.2888944"}],"container-title":["SIAM Journal on Imaging Sciences"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/epubs.siam.org\/doi\/pdf\/10.1137\/23M1565243","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T13:28:01Z","timestamp":1787318881000},"score":1,"resource":{"primary":{"URL":"https:\/\/epubs.siam.org\/doi\/10.1137\/23M1565243"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,1,18]]},"references-count":57,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2024,3,31]]}},"alternative-id":["10.1137\/23M1565243"],"URL":"https:\/\/doi.org\/10.1137\/23m1565243","relation":{},"ISSN":["1936-4954"],"issn-type":[{"value":"1936-4954","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,1,18]]}}}