Abstract
Solving ill-posed inverse problems can be done accurately if a regularizer well adapted to the nature of the data is available. Such regularizer can be systematically linked with the distribution of the data itself through the maximum a posteriori Bayesian framework. Recently, regularizers designed with the help of deep neural networks (DNN) received impressive success. Such regularizers are typically learned from large datasets. To reduce the computational burden of this task, we propose to adapt the compressive learning framework to the learning of regularizers parametrized by DNN. Our work shows the feasibility of batchless learning of regularizers from a compressed dataset. In order to achieve this, we propose an approximation of the compression operator that can be calculated explicitly for the task of learning a regularizer by DNN. We show that the proposed regularizer is capable of modeling complex regularity prior and can be used for denoising.
This work was partly funded by ANR project EFFIREG - ANR-20-CE40-0001.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Chatalic, A., Carratino, L., De Vito, E., Rosasco, L.: Mean Nyström embeddings for adaptive compressive learning. In: International Conference on Artificial Intelligence and Statistics, pp. 9869–9889. PMLR (2022)
Demoment, G.: Image reconstruction and restoration: overview of common estimation structures and problems. IEEE Trans. Acoust. Speech Signal Process. 37(12), 2024–2036 (1989)
Engel, J., et al.: Neural audio synthesis of musical notes with wavenet autoencoders. In: International Conference on Machine Learning, pp. 1068–1077. PMLR (2017)
Gribonval, R., Blanchard, G., Keriven, N., Traonmilin, Y.: Compressive statistical learning with random feature moments. Math. Stat. Learn. 3(2), 113–164 (2021)
Gribonval, R., Blanchard, G., Keriven, N., Traonmilin, Y.: Statistical learning guarantees for compressive clustering and compressive mixture modeling. Math. Stat. Learn. 3(2), 165–257 (2021)
Gribonval, R., Chatalic, A., Keriven, N., Schellekens, V., Jacques, L., Schniter, P.: Sketching datasets for large-scale learning (long version). arXiv preprint arXiv:2008.01839 (2020)
Gribonval, R., Chatalic, A., Keriven, N., Schellekens, V., Jacques, L., Schniter, P.: Sketching data sets for large-scale learning: keeping only what you need. IEEE Signal Process. Mag. 38(5), 12–36 (2021)
Hornik, K., Stinchcombe, M., White, H.: Multilayer feedforward networks are universal approximators. Neural Netw. 2(5), 359–366 (1989)
Hurault, S., Leclaire, A., Papadakis, N.: Gradient step denoiser for convergent plug-and-play. arXiv preprint arXiv:2110.03220 (2021)
Keriven, N., Bourrier, A., Gribonval, R., Pérez, P.: Sketching for large-scale learning of mixture models. Inf. Inference 7(3), 447–508 (2018)
Kobler, E., Effland, A., Kunisch, K., Pock, T.: Total deep variation: a stable regularization method for inverse problems. IEEE Trans. Pattern Anal. Mach. Intell. 44(12), 9163–9180 (2021)
Lunz, S., Öktem, O., Schönlieb, C.B.: Adversarial regularizers in inverse problems. In: Advances in Neural Information Processing Systems, vol. 31 (2018)
Nair, V., Hinton, G.E.: Rectified linear units improve restricted Boltzmann machines. In: International Conference on Machine Learning (2010)
Pan, X., Srikumar, V.: Expressiveness of rectifier networks. In: International Conference on Machine Learning, pp. 2427–2435. PMLR (2016)
Prost, J., Houdard, A., Almansa, A., Papadakis, N.: Learning local regularization for variational image restoration. In: Elmoataz, A., Fadili, J., Quéau, Y., Rabin, J., Simon, L. (eds.) SSVM 2021. LNCS, vol. 12679, pp. 358–370. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-75549-2_29
Schellekens, V.: Pycle: a python compressive learning toolbox (2020). https://doi.org/10.5281/zenodo.3855114
Schellekens, V., Jacques, L.: Compressive classification (machine learning without learning). arXiv preprint arXiv:1812.01410 (2018)
Schellekens, V., Jacques, L.: Compressive learning of generative networks. arXiv preprint arXiv:2002.05095 (2020)
Shi, H.: https://github.com/shihui1224/sketching-deep-regu-for-denoising
Shi, H., Traonmilin, Y., Aujol, J.F.: Compressive learning for patch-based image denoising. SIAM J. Imag. Sci. 15(3), 1184–1212 (2022)
Venkatakrishnan, S.V., Bouman, C.A., Wohlberg, B.: Plug-and-play priors for model based reconstruction. In: IEEE Global Conference on Signal and Information Processing, pp. 945–948. IEEE (2013)
Zoran, D., Weiss, Y.: From learning models of natural image patches to whole image restoration. In: International Conference on Computer Vision, pp. 479–486. IEEE (2011)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Shi, H., Traonmilin, Y., Aujol, JF. (2023). Compressive Learning of Deep Regularization for Denoising. In: Calatroni, L., Donatelli, M., Morigi, S., Prato, M., Santacesaria, M. (eds) Scale Space and Variational Methods in Computer Vision. SSVM 2023. Lecture Notes in Computer Science, vol 14009. Springer, Cham. https://doi.org/10.1007/978-3-031-31975-4_13
Download citation
DOI: https://doi.org/10.1007/978-3-031-31975-4_13
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-31974-7
Online ISBN: 978-3-031-31975-4
eBook Packages: Computer ScienceComputer Science (R0)