{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,20]],"date-time":"2026-11-20T19:47:28Z","timestamp":1795204048106,"version":"4.10.3"},"reference-count":71,"publisher":"Society for Industrial & Applied Mathematics (SIAM)","issue":"3","funder":[{"DOI":"10.13039\/501100001665","name":"Agence Nationale de la Recherche","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001665","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100009767","name":"Heriot-Watt University","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100009767","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004795","name":"Institut Universitaire de France","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100004795","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100000266","name":"Engineering and Physical Sciences Research Council","doi-asserted-by":"publisher","award":["EP\/T028270\/1"],"award-info":[{"award-number":["EP\/T028270\/1"]}],"id":[{"id":"10.13039\/501100000266","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["SIAM J. Imaging Sci."],"published-print":{"date-parts":[[2021,1]]},"abstract":"<jats:p>We introduce a new paradigm for solving regularized variational problems. These are typically formulated to address ill-posed inverse problems encountered in signal and image processing. The objective function is traditionally defined by adding a regularization function to a data fit term, which is subsequently minimized by using iterative optimization algorithms. Recently, several works have proposed to replace the operator related to the regularization by a more sophisticated denoiser. These approaches, known as plug-and-play (PnP) methods, have shown excellent performance. Although it has been noticed that, under some Lipschitz properties on the denoisers, the convergence of the resulting algorithm is guaranteed, little is known about characterizing the asymptotically delivered solution. In the current article, we propose to address this limitation. More specifically, instead of employing a functional regularization, we perform an operator regularization, where a maximally monotone operator (MMO) is learned in a supervised manner. This formulation is flexible as it allows the solution to be characterized through a broad range of variational inequalities, and it includes convex regularizations as special cases. From an algorithmic standpoint, the proposed approach consists in replacing the resolvent of the MMO by a neural network (NN). We present a universal approximation theorem proving that nonexpansive NNs are suitable models for the resolvent of a wide class of MMOs. The proposed approach thus provides a sound theoretical framework for analyzing the asymptotic behavior of first-order PnP algorithms. In addition, we propose a numerical strategy to train NNs corresponding to resolvents of MMOs. We apply our approach to image restoration problems and demonstrate its validity in terms of both convergence and quality.<\/jats:p>","DOI":"10.1137\/20m1387961","type":"journal-article","created":{"date-parts":[[2021,8,19]],"date-time":"2021-08-19T12:05:37Z","timestamp":1629374737000},"page":"1206-1237","source":"Crossref","is-referenced-by-count":126,"title":["Learning Maximally Monotone Operators for Image Recovery"],"prefix":"10.1137","volume":"14","author":[{"given":"Jean-Christophe","family":"Pesquet","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6296-6957","authenticated-orcid":true,"given":"Audrey","family":"Repetti","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Matthieu","family":"Terris","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yves","family":"Wiaux","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"351","published-online":{"date-parts":[[2021,8,19]]},"reference":[{"key":"atypb1","doi-asserted-by":"publisher","DOI":"10.1093\/mnras\/stz2117"},{"key":"atypb2","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2018.2799231"},{"key":"atypb3","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2019.2949470"},{"key":"atypb4","unstructured":"A. Alacaoglu, Y. Malitsky, and V. Cevher,\n                      Convergence of Adaptive Algorithms for Weakly Convex Constrained Optimization\n                      , preprint,https:\/\/arxiv.org\/abs\/2006.06650, 2020."},{"key":"atypb5","first-page":"291","author":"Anil C.","year":"2019","journal-title":"PMLR"},{"key":"atypb6","doi-asserted-by":"publisher","DOI":"10.1137\/18M1207685"},{"key":"atypb7","doi-asserted-by":"crossref","unstructured":"H. H. Bauschke and P. L. Combettes,\n                      Convex Analysis and Monotone Operator Theory in Hilbert Spaces\n                      , Springer, New York, 2017.","DOI":"10.1007\/978-3-319-48311-5"},{"key":"atypb8","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6420\/ab460a"},{"key":"atypb9","unstructured":"A. Bibi, B. Ghanem, V. Koltun, and R. Ranftl,\n                      Deep layers as stochastic solvers\n                      , in Proceedings of the International Conference on Learning Representations, 2019."},{"key":"atypb10","doi-asserted-by":"publisher","DOI":"10.1093\/mnras\/sty1182"},{"key":"atypb11","doi-asserted-by":"publisher","DOI":"10.1007\/s10107-020-01501-5"},{"key":"atypb12","doi-asserted-by":"publisher","DOI":"10.1137\/090769521"},{"key":"atypb13","doi-asserted-by":"publisher","DOI":"10.1109\/LSP.2013.2259813"},{"key":"atypb14","doi-asserted-by":"publisher","DOI":"10.1137\/S1052623495290179"},{"key":"atypb15","first-page":"9061","author":"Chen X.","year":"2018","journal-title":"Advances in Neural Information Processing Systems"},{"key":"atypb16","first-page":"854","author":"Cisse M.","year":"2017","journal-title":"JMLR"},{"key":"atypb17","unstructured":"R. Cohen, M. Elad, and P. Milanfar,\n                      Regularization by Denoising via Fixed-Point Projection (RED-PRO)\n                      , preprint,https:\/\/arxiv.org\/abs\/2008.00226, 2020."},{"key":"atypb18","doi-asserted-by":"publisher","DOI":"10.1007\/s10107-018-1303-3"},{"key":"atypb19","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2004.832922"},{"key":"atypb20","first-page":"185","author":"Combettes P. L.","year":"2011","journal-title":"New York"},{"key":"atypb21","doi-asserted-by":"publisher","DOI":"10.1007\/s11228-019-00526-z"},{"key":"atypb22","doi-asserted-by":"publisher","DOI":"10.1137\/19M1272780"},{"key":"atypb23","doi-asserted-by":"crossref","unstructured":"P. L. Combettes and J. C. Pesquet,\n                      Fixed point strategies in data science\n                      , IEEE Trans. Signal Process. (2021),https:\/\/doi.org\/10.1109\/TSP.2021.3069677.","DOI":"10.1109\/TSP.2021.3069677"},{"key":"atypb24","doi-asserted-by":"publisher","DOI":"10.1137\/050626090"},{"key":"atypb25","doi-asserted-by":"publisher","DOI":"10.1007\/s10957-012-0245-9"},{"key":"atypb26","first-page":"1806","author":"Condat L.","year":"2014","journal-title":"Proceedings of the IEEE European Signal Processing Conference"},{"key":"atypb27","doi-asserted-by":"publisher","DOI":"10.1002\/cpa.20042"},{"key":"atypb28","first-page":"248","author":"Deng J.","year":"2009","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"},{"key":"atypb29","first-page":"11427","author":"Fazlyab M.","year":"2019","journal-title":"Advances in Neural Information Processing Systems"},{"key":"atypb30","doi-asserted-by":"crossref","unstructured":"G. H. Golub and C. F. Van Loan,\n                      Matrix Computations\n                      , Vol. 3, JHU Press, Baltimore, 2013.","DOI":"10.56021\/9781421407944"},{"key":"atypb31","first-page":"5767","author":"Gulrajani I.","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"key":"atypb32","unstructured":"J. Hertrich, S. Neumayer, and G. Steidl,\n                      Convolutional Proximal Neural Networks and Plug-and-Play Algorithms\n                      , preprint,https:\/\/arxiv.org\/abs\/2011.02281, 2020."},{"key":"atypb33","unstructured":"J. Hoffman, D. A. Roberts, and S. Yaida,\n                      Robust Learning with Jacobian Regularization\n                      , preprint,https:\/\/arxiv.org\/abs\/1908.02729, 2019."},{"key":"atypb34","doi-asserted-by":"publisher","DOI":"10.1016\/0893-6080(89)90020-8"},{"key":"atypb35","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2021.3054506"},{"key":"atypb36","unstructured":"D. P. Kingma and J. Ba,\n                      Adam: A method for stochastic optimization\n                      , in Proceedings of the International Conference on Learning Representations, 2015."},{"key":"atypb37","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2014.2377273"},{"key":"atypb38","unstructured":"F. Latorre, P. Rolland, and V. Cevher,\n                      Lipschitz constant estimation of neural networks via sparse polynomial optimization\n                      , in Proceedings of the International Conference on Learning Representations, 2020."},{"key":"atypb39","first-page":"4681","author":"Ledig C.","year":"2017","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"},{"key":"atypb40","doi-asserted-by":"publisher","DOI":"10.1016\/S0893-6080(05)80131-5"},{"key":"atypb41","doi-asserted-by":"crossref","unstructured":"A. Levin, Y. Weiss, F. Durand, and W. Freeman,\n                      Understanding and evaluating blind deconvolution algorithms\n                      , in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2009.","DOI":"10.1109\/CVPR.2009.5206815"},{"key":"atypb42","doi-asserted-by":"crossref","unstructured":"D. G. Luenberger and Y. Ye,\n                      Linear and Nonlinear Programming,\n                      4th ed., Internat. Ser. Oper. Res. Management Sci. 228, Springer, New York, 2016.","DOI":"10.1007\/978-3-319-18842-3_6"},{"key":"atypb43","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2020.3014721"},{"key":"atypb44","unstructured":"S. Mallat,\n                      A Wavelet Tour of Signal Processing: The Sparse Way\n                      , Academic Press, New York, 2008."},{"key":"atypb45","first-page":"416","volume":"2","author":"Martin D.","year":"2001","journal-title":"Proceedings of the IEEE International Conference on Computer Vision"},{"key":"atypb46","first-page":"1781","author":"Meinhardt T.","year":"2017","journal-title":"Proceedings of the IEEE International Conference on Computer Vision"},{"key":"atypb47","unstructured":"T. Miyato, T. Kataoka, M. Koyama, and Y. Yoshida,\n                      Spectral normalization for generative adversarial networks\n                      , in Proceedings of the International Conference on Learning Representations, 2018."},{"key":"atypb48","doi-asserted-by":"publisher","DOI":"10.24033\/bsmf.1625"},{"key":"atypb49","first-page":"2545","author":"Mukkamala M. C.","year":"2017","journal-title":"JMLR"},{"key":"atypb50","first-page":"8389","author":"Neacsu A.","year":"2020","journal-title":"Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing"},{"key":"atypb51","unstructured":"H. Petzka, A. Fischer, and D. Lukovnikov,\n                      On the regularization of wasserstein gans\n                      , in Proceedings of the International Conference on Learning Representations, 2018."},{"key":"atypb52","doi-asserted-by":"publisher","DOI":"10.2140\/pjm.1966.17.497"},{"key":"atypb53","doi-asserted-by":"publisher","DOI":"10.1016\/0167-2789(92)90242-F"},{"key":"atypb54","first-page":"5546","author":"Ryu E.","year":"2019","journal-title":"PMLR"},{"key":"atypb55","first-page":"3839","volume":"31","author":"Scaman K.","year":"2018","journal-title":"Adv. Neural Inform. Process. Syst."},{"key":"atypb56","unstructured":"H. Sedghi, V. Gupta, and P. M. Long,\n                      The singular values of convolutional layers\n                      , in Proceedings of the International Conference on Learning Representations (ICLR), 2019."},{"key":"atypb57","unstructured":"C. Szegedy, W. Zaremba, I. Sutskever, J. Bruna, D. Erhan, I. J. Goodfellow, and R. Fergus,\n                      Intriguing properties of neural networks\n                      , in Proceedings of the International Conference on Learning Representations, 2014."},{"key":"atypb58","first-page":"8658","author":"Terris M.","year":"2020","journal-title":"Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing"},{"key":"atypb59","first-page":"3835","author":"Virmaux A.","year":"2018","journal-title":"Advances in Neural Information Processing Systems"},{"key":"atypb60","doi-asserted-by":"publisher","DOI":"10.1007\/s10444-011-9254-8"},{"key":"atypb61","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2003.819861"},{"key":"atypb62","unstructured":"X. Wei, B. Gong, Z. Liu, W. Lu, and L. Wang,\n                      Improving the improved training of Wasserstein GANS: A consistency term and its dual effect\n                      , in Proceedings of the International Conference on Learning Representations, 2018."},{"key":"atypb63","first-page":"1790","author":"Xu L.","year":"2014","journal-title":"Advances in Neural Information Processing Systems"},{"key":"atypb64","doi-asserted-by":"publisher","DOI":"10.1109\/LSP.2020.3006390"},{"key":"atypb65","unstructured":"Y. Yoshida and T. Miyato,\n                      Spectral Norm Regularization for Improving the Generalizability of Deep Learning\n                      , preprint,https:\/\/arxiv.org\/abs\/1705.10941, 2017."},{"key":"atypb66","first-page":"1828","author":"Zhang J.","year":"2018","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"},{"key":"atypb67","first-page":"1","volume":"1","author":"Zhang K.","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"atypb68","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2017.2662206"},{"key":"atypb69","first-page":"3929","author":"Zhang K.","year":"2017","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"},{"key":"atypb70","first-page":"3262","author":"Zhang K.","year":"2018","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"},{"key":"atypb71","first-page":"1671","author":"Zhang K.","year":"2019","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"}],"container-title":["SIAM Journal on Imaging Sciences"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/epubs.siam.org\/doi\/pdf\/10.1137\/20M1387961","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T13:53:11Z","timestamp":1787320391000},"score":1,"resource":{"primary":{"URL":"https:\/\/epubs.siam.org\/doi\/10.1137\/20M1387961"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,1]]},"references-count":71,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2021,1]]}},"alternative-id":["10.1137\/20M1387961"],"URL":"https:\/\/doi.org\/10.1137\/20m1387961","relation":{},"ISSN":["1936-4954"],"issn-type":[{"value":"1936-4954","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,1]]}}}