{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T01:02:23Z","timestamp":1760058143690,"version":"build-2065373602"},"reference-count":34,"publisher":"MDPI AG","issue":"3","license":[{"start":{"date-parts":[[2025,3,12]],"date-time":"2025-03-12T00:00:00Z","timestamp":1741737600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100004410","name":"Turkish Scientific and Technical Research Council (TUBITAK) under the BIDEB-2232 program","doi-asserted-by":"publisher","award":["118C203"],"award-info":[{"award-number":["118C203"]}],"id":[{"id":"10.13039\/501100004410","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Symmetry"],"abstract":"<jats:p>We develop a new unsupervised symmetry learning method that starts with raw data and provides the minimal generator of an underlying Lie group of symmetries, together with a symmetry-equivariant representation of the data, which turns the hidden symmetry into an explicit one. The method is able to learn the pixel translation operator from a dataset with only an approximate translation symmetry and can learn quite different types of symmetries that are not apparent to the naked eye. The method is based on the formulation of an information-theoretic loss function that measures both the degree of symmetry of a dataset under a candidate symmetry generator and a proposed notion of locality of the samples, which is coupled to symmetry. We demonstrate that this coupling between symmetry and locality, together with an optimization technique developed for entropy estimation, results in a stable system that provides reproducible results.<\/jats:p>","DOI":"10.3390\/sym17030425","type":"journal-article","created":{"date-parts":[[2025,3,12]],"date-time":"2025-03-12T10:45:06Z","timestamp":1741776306000},"page":"425","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["SymmetryLens: Unsupervised Symmetry Learning via Locality and Density Preservation"],"prefix":"10.3390","volume":"17","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-1619-2776","authenticated-orcid":false,"given":"Onur","family":"Efe","sequence":"first","affiliation":[{"name":"Department of Physics, Bogazici University, Istanbul 34342, Turkey"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2318-9227","authenticated-orcid":false,"given":"Arkadas","family":"Ozakin","sequence":"additional","affiliation":[{"name":"Department of Physics, Bogazici University, Istanbul 34342, Turkey"}]}],"member":"1968","published-online":{"date-parts":[[2025,3,12]]},"reference":[{"key":"ref_1","unstructured":"Cohen, T., and Welling, M. (2016, January 19\u201324). Group equivariant convolutional networks. Proceedings of the International Conference on Machine Learning. PMLR, New York, NY, USA."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Babelon, O., Bernard, D., and Talon, M. (2003). Introduction to Classical Integrable Systems, Cambridge Monographs on Mathematical Physics, Cambridge University Press.","DOI":"10.1017\/CBO9780511535024"},{"key":"ref_3","unstructured":"Hairer, E., Lubich, C., and Wanner, G. (2010). Geometric Numerical Integration: Structure-Preserving Algorithms for Ordinary Differential Equations, Springer. [2nd ed.]."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Higgins, I., Racani\u00e8re, S., and Rezende, D. (2022). Symmetry-based representations for artificial and biological general intelligence. Front. Comput. Neurosci., 16.","DOI":"10.3389\/fncom.2022.836498"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Anselmi, F., and Patel, A.B. (2022). Symmetry as a guiding principle in artificial and brain neural networks. Front. Comput. Neurosci., 16.","DOI":"10.3389\/fncom.2022.1039572"},{"key":"ref_6","unstructured":"Bronstein, M.M., Bruna, J., Cohen, T., and Veli\u010dkovi\u0107, P. (2021). Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges. arXiv."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"201","DOI":"10.1016\/j.patcog.2018.07.025","article-title":"Symmetry-adapted representation learning","volume":"86","author":"Anselmi","year":"2019","journal-title":"Pattern Recognit."},{"key":"ref_8","unstructured":"Helgason, S. (1979). Differential Geometry, Lie Groups, and Symmetric Spaces, Academic Press."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"096031","DOI":"10.1103\/PhysRevD.105.096031","article-title":"Symmetry discovery with deep learning","volume":"105","author":"Desai","year":"2022","journal-title":"Phys. Rev. D"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Ma, Y., and Fu, Y. (2011). Density-Preserving Maps. Manifold Learning: Theory and Applications, CRC Press.","DOI":"10.1201\/b11431"},{"key":"ref_11","unstructured":"Weinberg, S. (1997). What is Quantum Field Theory, and What Did We Think It Is?. arXiv."},{"key":"ref_12","first-page":"17605","article-title":"Learning invariances in neural networks from training data","volume":"33","author":"Benton","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_13","unstructured":"Romero, D.W., and Lohit, S. (2021). Learning partial equivariances from data. arXiv."},{"key":"ref_14","unstructured":"Zhou, A., Knowles, T., and Finn, C. (2020). Meta-learning symmetries by reparameterization. arXiv."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"096030","DOI":"10.1103\/PhysRevD.105.096030","article-title":"Machine learning a manifold","volume":"105","author":"Craven","year":"2022","journal-title":"Phys. Rev. D"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"138266","DOI":"10.1016\/j.physletb.2023.138266","article-title":"Accelerated discovery of machine-learned symmetries: Deriving the exceptional Lie groups G2, F4 and E6","volume":"847","author":"Forestano","year":"2023","journal-title":"Phys. Lett. B"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"025027","DOI":"10.1088\/2632-2153\/acd989","article-title":"Deep learning symmetries and their Lie groups, algebras, and subalgebras from first principles","volume":"4","author":"Forestano","year":"2023","journal-title":"Mach. Learn. Sci. Technol."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"015010","DOI":"10.1088\/2632-2153\/abbd2d","article-title":"Detecting symmetries with neural networks","volume":"2","author":"Krippendorf","year":"2020","journal-title":"Mach. Learn. Sci. Technol."},{"key":"ref_19","unstructured":"Sohl-Dickstein, J., Wang, C.M., and Olshausen, B.A. (2010). An unsupervised algorithm for learning lie group transformations. arXiv."},{"key":"ref_20","first-page":"2503","article-title":"Automatic symmetry discovery with lie algebra convolutional network","volume":"34","author":"Dehmamy","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_21","first-page":"15353","article-title":"Hamiltonian neural networks","volume":"32","author":"Greydanus","year":"2019","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_22","first-page":"16384","article-title":"Noether networks: Meta-learning useful conserved quantities","volume":"34","author":"Alet","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_23","unstructured":"Goodfellow, I.J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y. (2014). Generative Adversarial Networks. arXiv."},{"key":"ref_24","unstructured":"Yang, J., Walters, R., Dehmamy, N., and Yu, R. (2023, January 23\u201329). Generative adversarial symmetry discovery. Proceedings of the International Conference on Machine Learning. PMLR, Honolulu, HI, USA."},{"key":"ref_25","unstructured":"Yang, J., Dehmamy, N., Walters, R., and Yu, R. (2023). Latent Space Symmetry Discovery. arXiv."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"P08024","DOI":"10.1088\/1748-0221\/17\/08\/P08024","article-title":"A method to challenge symmetries in data with self-supervised learning","volume":"17","author":"Tombs","year":"2022","journal-title":"J. Instrum."},{"key":"ref_27","unstructured":"Doob, J. (1991). Stochastic Processes, Wiley."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Folland, G.B. (2016). A Course in Abstract Harmonic Analysis, CRC Press.","DOI":"10.1201\/b19172"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"1129","DOI":"10.1162\/neco.1995.7.6.1129","article-title":"An information-maximization approach to blind separation and blind deconvolution","volume":"7","author":"Bell","year":"1995","journal-title":"Neural Comput."},{"key":"ref_30","unstructured":"Cover, T. (1999). Elements of Information Theory, Wiley-India."},{"key":"ref_31","unstructured":"Pichler, G., Colombo, P.J.A., Boudiaf, M., Koliander, G., and Piantanida, P. (2022, January 17\u201323). A differential entropy estimator for training neural networks. Proceedings of the International Conference on Machine Learning. PMLR, Baltimore, MD, USA."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Herman, J., and Usher, W. (2017). SALib: An open-source Python library for Sensitivity Analysis. J. Open Source Softw., 2.","DOI":"10.21105\/joss.00097"},{"key":"ref_33","first-page":"18155","article-title":"Toward SALib 2.0: Advancing the accessibility and interpretability of global sensitivity analyses","volume":"4","author":"Iwanaga","year":"2022","journal-title":"Socio-Environ. Syst. Model."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"161","DOI":"10.1080\/00401706.1991.10484804","article-title":"Factorial sampling plans for preliminary computational experiments","volume":"33","author":"Morris","year":"1991","journal-title":"Technometrics"}],"container-title":["Symmetry"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2073-8994\/17\/3\/425\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T16:51:24Z","timestamp":1760028684000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2073-8994\/17\/3\/425"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3,12]]},"references-count":34,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2025,3]]}},"alternative-id":["sym17030425"],"URL":"https:\/\/doi.org\/10.3390\/sym17030425","relation":{},"ISSN":["2073-8994"],"issn-type":[{"type":"electronic","value":"2073-8994"}],"subject":[],"published":{"date-parts":[[2025,3,12]]}}}