{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T00:57:19Z","timestamp":1773795439694,"version":"3.50.1"},"reference-count":52,"publisher":"MIT Press","issue":"4","content-domain":{"domain":["direct.mit.edu"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,3,17]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>We introduce the cooperative network architecture (CNA), a model that represents sensory signals using structured, recurrently connected networks of neurons, termed \u201cnets.\u201d Nets are dynamically assembled from overlapping net fragments, which are learned based on statistical regularities in sensory input. This architecture offers robustness to noise, deformation, and generalization to out-of-distribution data, addressing challenges in current vision systems from a novel perspective. We demonstrate that net fragments can be learned without supervision and flexibly recombined to encode novel patterns, enabling figure\u00a0completion and resilience to noise. Our findings establish CNA as a promising paradigm for developing neural representations that integrate local feature processing with global structure formation, providing a foundation for future research on invariant object recognition.<\/jats:p>","DOI":"10.1162\/neco.a.1505","type":"journal-article","created":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T17:51:16Z","timestamp":1772733076000},"page":"538-572","update-policy":"https:\/\/doi.org\/10.1162\/mitpressjournals.corrections.policy","source":"Crossref","is-referenced-by-count":1,"title":["The Cooperative Network Architecture: Learning Structured Networks as Representation of Sensory Patterns"],"prefix":"10.1162","volume":"38","author":[{"given":"Pascal J.","family":"Sager","sequence":"first","affiliation":[{"name":"Centre for Artificial Intelligence, Zurich University of Applied Sciences, 8400 Winterthur, Switzerland"},{"name":"Institute of Neuroinformatics, University of Zurich and ETH Zurich, 8057 Zurich, Switzerland sage@zhaw.ch"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jan M.","family":"Deriu","sequence":"additional","affiliation":[{"name":"Centre for Artificial Intelligence, Zurich University of Applied Sciences, 8400 Winterthur, Switzerland deri@zhaw.ch"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Benjamin F.","family":"Grewe","sequence":"additional","affiliation":[{"name":"Institute of Neuroinformatics, University of Zurich and ETH Zurich, Zurich 8057, Switzerland bgrewe@ethz.ch"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Thilo","family":"Stadelmann","sequence":"additional","affiliation":[{"name":"Centre for Artificial Intelligence, Zurich University of Applied Sciences, 8400 Winterthur, Switzerland"},{"name":"European Centre for Living Technology, 30123 Venice, Italy stdm@zhaw.ch"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Christoph von der","family":"Malsburg","sequence":"additional","affiliation":[{"name":"Centre for Artificial Intelligence, Zurich University of Applied Sciences, 8400 Winterthur, Switzerland"},{"name":"Institute of Neuroinformatics, University of Zurich and ETH Zurich, Zurich 8057, Switzerland"},{"name":"Frankfurt Institute for Advanced Studies, 60438 Frankfurt, Germany malsburg@fias.uni-frankfurt.de"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"281","published-online":{"date-parts":[[2026,3,17]]},"reference":[{"key":"2026031717004593900_bib1","author":"Ahmad","year":"2015","journal-title":"Properties of sparse distributed representations and their application to hierarchical temporal memory"},{"issue":"11","key":"2026031717004593900_bib2","doi-asserted-by":"publisher","first-page":"1197","DOI":"10.1109\/T-C.1972.223477","article-title":"Learning patterns and pattern sequences by self-organizing nets of threshold elements","volume":"C-21","author":"Amari","year":"1972","journal-title":"IEEE Transactions on Computers"},{"key":"2026031717004593900_bib3","doi-asserted-by":"publisher","first-page":"346","DOI":"10.1007\/978-3-319-99978-4_27","article-title":"Trace and detect adversarial attacks on CNNs using feature response maps","volume-title":"Lecture Notes in Computer Science","author":"Amirian","year":"2018"},{"issue":"17","key":"2026031717004593900_bib4","doi-asserted-by":"publisher","first-page":"6297","DOI":"10.1073\/pnas.84.17.6297","article-title":"Shifter circuits: A computational strategy for dynamic aspects of visual processing","volume":"84","author":"Anderson","year":"1987","journal-title":"Proceedings of the National Academy of Sciences"},{"key":"2026031717004593900_bib5","volume-title":"Map-seeking circuits in visual cognition: A computational mechanism for biological and machine vision","author":"Arathorn","year":"2002"},{"key":"2026031717004593900_bib6","first-page":"3227","article-title":"Hallucinating agnostic images to generalize across domains","volume-title":"Proceedings of the IEEE\/CVF International Conference on Computer Vision Workshops","author":"Carlucci","year":"2019"},{"key":"2026031717004593900_bib7","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-319-58347-1","volume-title":"Domain adaptation in computer vision applications","author":"Csurka","year":"2017"},{"key":"2026031717004593900_bib8","doi-asserted-by":"crossref","first-page":"248","DOI":"10.1109\/CVPR.2009.5206848","article-title":"ImageNet: A large-scale hierarchical image database","volume-title":"Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition","author":"Deng","year":"2009"},{"key":"2026031717004593900_bib9","article-title":"Towards robust object detection invariant to real-world domain shifts","volume-title":"Proceedings of the International Conference on Learning Representations","author":"Fan","year":"2023"},{"issue":"1","key":"2026031717004593900_bib10","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1016\/0010-0277(88)90031-5","article-title":"Connectionism and cognitive architecture: A critical analysis","volume":"28","author":"Fodor","year":"1988","journal-title":"Cognition"},{"issue":"1","key":"2026031717004593900_bib11","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1162\/neco.1992.4.1.1","article-title":"Neural networks and the bias\/variance dilemma","volume":"4","author":"Geman","year":"1992","journal-title":"Neural Computation"},{"key":"2026031717004593900_bib12","author":"Greff","year":"2020","journal-title":"On the binding problem in artificial neural networks."},{"key":"2026031717004593900_bib13","volume-title":"The sensory order: An inquiry into the foundations of theoretical psychology","author":"Hayek","year":"1952"},{"key":"2026031717004593900_bib14","volume-title":"The organization of behavior: A neuropsychological theory","author":"Hebb","year":"1949"},{"issue":"5786","key":"2026031717004593900_bib15","doi-asserted-by":"publisher","first-page":"504","DOI":"10.1126\/science.1127647","article-title":"Reducing the dimensionality of data with neural networks","volume":"313","author":"Hinton","year":"2006","journal-title":"Science"},{"key":"2026031717004593900_bib16","first-page":"448","article-title":"Optimal perceptual inference","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"Hinton","year":"1983"},{"key":"2026031717004593900_bib17","author":"Hinton","year":"1984","journal-title":"Boltzmann machines: Constraint satisfaction networks that learn"},{"issue":"8","key":"2026031717004593900_bib18","doi-asserted-by":"publisher","first-page":"2554","DOI":"10.1073\/pnas.79.8.2554","article-title":"Neural networks and physical systems with emergent collective computational abilities","volume":"79","author":"Hopfield","year":"1982","journal-title":"Proceedings of the National Academy of Sciences"},{"key":"2026031717004593900_bib19","article-title":"Adam: A method for stochastic optimization","volume-title":"Proceedings of the International Conference on Learning Representations.","author":"Kingma","year":"2015"},{"key":"2026031717004593900_bib20","author":"Krizhevsky","year":"2009","journal-title":"Learning multiple layers of features from tiny images."},{"key":"2026031717004593900_bib21","doi-asserted-by":"publisher","first-page":"258","DOI":"10.1016\/j.neunet.2022.07.003","article-title":"Hierarchical binding in convolutional neural networks: Making adversarial attacks geometrically challenging","volume":"155","author":"Leadholm","year":"2022","journal-title":"Neural Networks"},{"issue":"7553","key":"2026031717004593900_bib22","doi-asserted-by":"publisher","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"LeCun","year":"2015","journal-title":"Nature"},{"issue":"4","key":"2026031717004593900_bib23","doi-asserted-by":"publisher","first-page":"541","DOI":"10.1162\/neco.1989.1.4.541","article-title":"Backpropagation applied to handwritten zip code recognition","volume":"1","author":"LeCun","year":"1989","journal-title":"Neural Computation"},{"issue":"11","key":"2026031717004593900_bib24","doi-asserted-by":"publisher","first-page":"2278","DOI":"10.1109\/5.726791","article-title":"Gradient-based learning applied to document recognition","volume":"86","author":"LeCun","year":"1998","journal-title":"Proceedings of the IEEE"},{"issue":"6","key":"2026031717004593900_bib25","doi-asserted-by":"publisher","first-page":"1473","DOI":"10.1162\/neco.2010.01-09-953","article-title":"Learning to represent spatial transformations with factored higher-order Boltzmann machines","volume":"22","author":"Memisevic","year":"2010","journal-title":"Neural Computation"},{"key":"2026031717004593900_bib26","doi-asserted-by":"crossref","DOI":"10.1109\/SDS66131.2025.00008","article-title":"Hounsfield unit ranges as inductive bias for intra-clinical learning of data-efficient CT segmentation models","volume-title":"Proceedings of the 12th IEEE Swiss Conference on Data Science","author":"Meyer","year":"2025"},{"key":"2026031717004593900_bib27","author":"Miconi","year":"2021","journal-title":"Hebbian learning with gradients: Hebbian convolutional neural networks with modern deep learning frameworks"},{"key":"2026031717004593900_bib28","volume-title":"Machine learning","author":"Mitchell","year":"1997"},{"key":"2026031717004593900_bib29","first-page":"2574","article-title":"DeepFool: A simple and accurate method to fool deep neural networks","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"Moosavi-Dezfooli","year":"2016"},{"key":"2026031717004593900_bib30","doi-asserted-by":"publisher","first-page":"45","DOI":"10.1007\/BF00962707","article-title":"A multiscale dynamic routing circuit for forming size- and position-invariant object representations","volume":"2","author":"Olshausen","year":"1995","journal-title":"Journal of Computational Neuroscience"},{"issue":"8","key":"2026031717004593900_bib31","doi-asserted-by":"publisher","first-page":"1665","DOI":"10.1162\/0899766054026639","article-title":"How close are we to understanding V1?","volume":"17","author":"Olshausen","year":"2005","journal-title":"Neural Computation"},{"key":"2026031717004593900_bib32","doi-asserted-by":"publisher","first-page":"165","DOI":"10.1016\/j.neunet.2012.08.013","article-title":"Neural associative memories and sparse coding","volume":"37","author":"Palm","year":"2013","journal-title":"Neural Networks"},{"key":"2026031717004593900_bib33","volume-title":"Understanding deep learning","author":"Prince","year":"2023"},{"key":"2026031717004593900_bib34","volume-title":"Principles of neurodynamics: Perceptrons and the theory of brain mechanisms","author":"Rosenblatt","year":"1962"},{"key":"2026031717004593900_bib35","author":"Sager","year":"2025","journal-title":"AI agents for computer use: A review of instruction-based computer control, GUI automation, and operator assistants"},{"issue":"8","key":"2026031717004593900_bib36","doi-asserted-by":"publisher","first-page":"222","DOI":"10.3390\/jimaging8080222","article-title":"Unsupervised domain adaptation for vertebrae detection and identification in 3D CT volumes using a domain sanity loss","volume":"8","author":"Sager","year":"2022","journal-title":"Journal of Imaging"},{"key":"2026031717004593900_bib37","first-page":"26","article-title":"A survey of un-, weakly-, and semi-supervised learning methods for noisy, missing and partial labels in industrial vision applications","author":"Simmler","year":"2021","journal-title":"Proceedings of the 8th Swiss Conference on Data Science"},{"key":"2026031717004593900_bib38","first-page":"194","article-title":"Information processing in dynamical systems: Foundations of harmony theory","volume-title":"Parallel distributed processing: Explorations in the microstructure of cognition","author":"Smolensky","year":"1987"},{"issue":"1","key":"2026031717004593900_bib39","article-title":"The enhanced storage capacity in neural networks with low activity level","volume":"6","author":"Tsodyks","year":"2007","journal-title":"Europhysics Letters"},{"issue":"1","key":"2026031717004593900_bib40","doi-asserted-by":"publisher","first-page":"1","DOI":"10.5334\/tismir.157","article-title":"Real world music object recognition","volume":"7","author":"Tuggener","year":"2024","journal-title":"Transactions of the International Society for Music Information Retrieval"},{"key":"2026031717004593900_bib41","first-page":"6000","article-title":"Attention is all you need","volume-title":"Advances in neural information processing systems","author":"Vaswani","year":"2017"},{"key":"2026031717004593900_bib42","author":"von der Malsburg","year":"1981","journal-title":"The correlation theory of brain function."},{"issue":"4","key":"2026031717004593900_bib43","doi-asserted-by":"publisher","first-page":"511","DOI":"10.17791\/jcs.2018.19.4.511","article-title":"Concerning the neuronal code","volume":"19","author":"von der Malsburg","year":"2018","journal-title":"Journal of Cognitive Science"},{"key":"2026031717004593900_bib44","volume-title":"Fodor and Pylyshyn\u2019s critique of connectionism and the brain as basis of the mind","author":"von der Malsburg","year":"2023"},{"issue":"1","key":"2026031717004593900_bib45","doi-asserted-by":"publisher","first-page":"67","DOI":"10.1007\/s10339-024-01214-z","article-title":"How are segmentation and binding computed and represented in the brain?","volume":"25","author":"von der Malsburg","year":"2024","journal-title":"Cognitive Processing"},{"key":"2026031717004593900_bib46","author":"von der Malsburg","year":"2022","journal-title":"A theory of natural intelligence."},{"issue":"6","key":"2026031717004593900_bib47","doi-asserted-by":"publisher","first-page":"1172","DOI":"10.1037\/a0029333","article-title":"A century of Gestalt psychology in visual perception: I. Perceptual grouping and figure-ground organization","volume":"138","author":"Wagemans","year":"2012","journal-title":"Psychological Bulletin"},{"issue":"7","key":"2026031717004593900_bib48","doi-asserted-by":"publisher","first-page":"775","DOI":"10.1109\/34.598235","article-title":"Face recognition by elastic bunch graph matching","volume":"19","author":"Wiskott","year":"1997","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"issue":"7","key":"2026031717004593900_bib49","doi-asserted-by":"publisher","first-page":"34","DOI":"10.1167\/8.7.34","article-title":"A recurrent dynamic model for correspondence-based face recognition","volume":"8","author":"Wolfrum","year":"2008","journal-title":"Journal of Vision"},{"issue":"7","key":"2026031717004593900_bib50","doi-asserted-by":"publisher","first-page":"1341","DOI":"10.1162\/neco.1996.8.7.1341","article-title":"The lack of a priori distinctions between learning algorithms","volume":"8","author":"Wolpert","year":"1996","journal-title":"Neural Computation"},{"key":"2026031717004593900_bib51","author":"Zeiler","year":"2013","journal-title":"Visualizing and understanding convolutional networks."},{"issue":"8","key":"2026031717004593900_bib52","doi-asserted-by":"publisher","first-page":"1311","DOI":"10.1016\/j.neunet.2004.06.010","article-title":"Maplets for correspondence-based object recognition","volume":"17","author":"Zhu","year":"2004","journal-title":"Neural Networks"}],"container-title":["Neural Computation"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/direct.mit.edu\/neco\/article-pdf\/38\/4\/538\/2585609\/neco.a.1505.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/direct.mit.edu\/neco\/article-pdf\/38\/4\/538\/2585609\/neco.a.1505.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,17]],"date-time":"2026-03-17T21:00:56Z","timestamp":1773781256000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/neco\/article\/38\/4\/538\/135632\/The-Cooperative-Network-Architecture-Learning"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,17]]},"references-count":52,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2026,3,17]]},"published-print":{"date-parts":[[2026,3,17]]}},"URL":"https:\/\/doi.org\/10.1162\/neco.a.1505","relation":{},"ISSN":["0899-7667","1530-888X"],"issn-type":[{"value":"0899-7667","type":"print"},{"value":"1530-888X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,4]]},"published":{"date-parts":[[2026,3,17]]}}}