{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,3]],"date-time":"2026-05-03T02:31:23Z","timestamp":1777775483078,"version":"3.51.4"},"reference-count":117,"publisher":"Emerald","issue":"5-6","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2019,11,28]]},"abstract":"<jats:p>Spectral methods have been the mainstay in several domains such as machine learning, applied mathematics and scientific computing. They involve finding a certain kind of spectral decomposition to obtain basis functions that can capture important structures or directions for the problem at hand. The most common spectral method is the principal component analysis (PCA). It utilizes the principal components or the top eigenvectors of the data covariance matrix to carry out dimensionality reduction as one of its applications. This data pre-processing step is often effective in separating signal from noise.<\/jats:p>\n                  <jats:p>PCA and other spectral techniques applied to matrices have several limitations. By limiting to only pairwise moments, they are effectively making a Gaussian approximation on the underlying data. Hence, they fail on data with hidden variables which lead to non-Gaussianity. However, in almost any data set, there are latent effects that cannot be directly observed, e.g., topics in a document corpus, or underlying causes of a disease. By extending the spectral decomposition methods to higher order moments, we demonstrate the ability to learn a wide range of latent variable models efficiently. Higher-order moments can be represented by tensors, and intuitively, they can encode more information than just pairwise moment matrices. More crucially, tensor decomposition can pick up latent effects that are missed by matrix methods. For instance, tensor decomposition can uniquely identify non-orthogonal components. Exploiting these aspects turns out to be fruitful for provable unsupervised learning of a wide range of latent variable models.<\/jats:p>\n                  <jats:p>We also outline the computational techniques to design efficient tensor decomposition methods. They are embarrassingly parallel and thus scalable to large data sets. Whilst there exist many optimized linear algebra software packages, efficient tensor algebra packages are also beginning to be developed. We introduce Tensorly, which has a simple python interface for expressing tensor operations. It has a flexible back-end system supporting NumPy, PyTorch, TensorFlow and MXNet amongst others. This allows it to carry out multi-GPU and CPU operations, and can also be seamlessly integrated with deep-learning functionalities.<\/jats:p>","DOI":"10.1561\/2200000057","type":"journal-article","created":{"date-parts":[[2019,11,28]],"date-time":"2019-11-28T10:02:06Z","timestamp":1574935326000},"page":"393-536","source":"Crossref","is-referenced-by-count":21,"title":["Spectral Learning on Matrices and Tensors"],"prefix":"10.1108","volume":"12","author":[{"given":"Majid","family":"Janzamin","sequence":"first","affiliation":[{"name":"Twitter"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Rong","family":"Ge","sequence":"additional","affiliation":[{"name":"Duke University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jean","family":"Kossaifi","sequence":"additional","affiliation":[{"name":"Imperial College London"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Anima","family":"Anandkumar","sequence":"additional","affiliation":[{"name":"NVIDIA and California Institute of Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"140","published-online":{"date-parts":[[2019,11,28]]},"reference":[{"key":"2026033012250342600_ref001","doi-asserted-by":"crossref","first-page":"256","DOI":"10.1007\/11427995_21","volume-title":"Intelligence and Security Informatics","author":"Acar","year":"2005"},{"key":"2026033012250342600_ref002","volume-title":"\u201cWhat regularized auto-encoders learn from the data generating distribution\u201d","author":"Alain","year":"2012"},{"key":"2026033012250342600_ref003","volume-title":"\u201cDoubly accelerated methods for faster CCA and generalized eigen decomposition\u201d","author":"Allen-Zhu","year":"2016"},{"key":"2026033012250342600_ref004","volume-title":"\u201cFirst Efficient Convergence for Streaming k-PCA: a Global, Gap-Free, and Near-Optimal Rate\u201d","author":"Allen-Zhu","year":"2016"},{"key":"2026033012250342600_ref005","volume-title":"Conference on Learning Theory (COLT)","author":"Anandkumar","year":"2013"},{"key":"2026033012250342600_ref006","first-page":"2773","article-title":"\"Tensor Methods for Learning Latent Variable Models\"","volume":"15","author":"Anandkumar","year":"2014","journal-title":"J. of Machine Learning Research"},{"key":"2026033012250342600_ref007","volume-title":"Conference on Learning Theory","author":"Anandkumar","year":"2017"},{"key":"2026033012250342600_ref008","volume-title":"Advances in Neural Information Processing Systems 25","author":"Anandkumar","year":"2012"},{"key":"2026033012250342600_ref009","volume-title":"\u201cSample Complexity Analysis for Learning Overcomplete Latent Variable Models through Tensor Methods\u201d","author":"\"Anandkumar","year":"2014"},{"key":"2026033012250342600_ref010","volume-title":"Advances in Neural Information Processing Systems 25","author":"Anandkumar","year":"2012"},{"key":"2026033012250342600_ref011","volume-title":"\u201cA method of moments for mixture models and hidden Markov models\u201d","author":"Anandkumar","year":"2012"},{"issue":"13","key":"2026033012250342600_ref012","doi-asserted-by":"crossref","first-page":"2053","DOI":"10.1021\/ac00236a025","article-title":"\u201cStrategies for analyzing data from video fluorometric monitoring of liquid chromatographic effluents\u201d","volume":"53","author":"Appellof","year":"1981","journal-title":"Analytical Chemistry"},{"key":"2026033012250342600_ref013","volume-title":"Proceedings of the forty-nineth annual ACM symposium on Theory of computing","author":"Arora","year":"2017"},{"key":"2026033012250342600_ref014","volume-title":"CoRR","author":"Astrid","year":"2017"},{"key":"2026033012250342600_ref015","doi-asserted-by":"crossref","first-page":"80","DOI":"10.1214\/08-PS124","article-title":"\u201cOn exchangeable random variables and the statistics of large graphs and hypergraphs\u201d","volume":"5","author":"Austin","year":"2008","journal-title":"Probab. Survey"},{"key":"2026033012250342600_ref016","first-page":"417","volume-title":"Conference on Learning Theory","author":"Barak","year":"2016"},{"key":"2026033012250342600_ref017","first-page":"5236","volume-title":"\u201cSum-of-squares proofs and the quest toward optimal algorithms\u201d","author":"Barak","year":"2014"},{"issue":"6","key":"2026033012250342600_ref018","doi-asserted-by":"crossref","first-page":"1554","DOI":"10.1214\/aoms\/1177699147","article-title":"\u201cStatistical inference for probabilistic functions of finite state Markov chains\u201d","volume":"37","author":"Baum","year":"1966","journal-title":"The annals of mathematical statistics"},{"key":"2026033012250342600_ref019","doi-asserted-by":"crossref","first-page":"594","DOI":"10.1145\/2591796.2591881","volume-title":"Proceedings of the forty-sixth annual ACM symposium on Theory of computing","author":"Bhaskara","year":"2014"},{"key":"2026033012250342600_ref020","volume-title":"Application of calculus of matrices to method of least squares: with special reference to geodetic calculations","author":"Bjerhammar","year":"1951"},{"key":"2026033012250342600_ref021","first-page":"993","article-title":"\u201cLatent dirichlet allocation\u201d","volume":"3","author":"Blei","year":"2003","journal-title":"Journal of machine Learning research"},{"key":"2026033012250342600_ref022","volume-title":"Vorabversion eines Lehrbuchs","author":"Blum","year":"2016"},{"key":"2026033012250342600_ref023","first-page":"93","volume-title":"IEEE International Symposium on Circuits and Systems","author":"Cardoso","year":"1996"},{"issue":"3","key":"2026033012250342600_ref024","doi-asserted-by":"crossref","first-page":"283","DOI":"10.1007\/BF02310791","article-title":"\u201cAnalysis of individual differences in multidimensional scaling via an N-way generalization of \u201cEckart- Young\u201d decomposition\u201d","volume":"35","author":"Carroll","year":"1970","journal-title":"Psychometrika"},{"key":"2026033012250342600_ref025","doi-asserted-by":"crossref","first-page":"51","DOI":"10.1016\/S0025-5564(96)00075-2","article-title":"\"Full reconstruction of Markov models on evolutionary trees: Identifiability and consistency\u201d","volume":"137","author":"Chang","year":"1996","journal-title":"Mathematical Biosciences"},{"issue":"3","key":"2026033012250342600_ref026","first-page":"287","volume":"36","author":"Comon","year":"1994","journal-title":"\u201cIndependent Component Analysis, a new concept?\u201d"},{"key":"2026033012250342600_ref027","first-page":"1","volume-title":"Mathematics in Signal Processing V","author":"Comon","year":"2002"},{"key":"2026033012250342600_ref028","volume-title":"Handbook of Blind Source Separation: Independent Component Analysis and Applications","author":"Comon","year":"2010"},{"issue":"1","key":"2026033012250342600_ref029","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1137\/0707001","article-title":"\u201cThe rotation of eigenvectors by a perturbation. III\u201d","volume":"7","author":"Davis","year":"1970","journal-title":"SIAM Journal on Numerical Analysis"},{"issue":"6","key":"2026033012250342600_ref030","doi-asserted-by":"crossref","first-page":"2965","DOI":"10.1109\/TSP.2007.893943","article-title":"\u201cFourth-order cumulant-based blind identification of underdetermined mixtures\u201d","volume":"55","author":"De Lathauwer","year":"2007","journal-title":"Signal Processing, IEEE Transactions on"},{"key":"2026033012250342600_ref031","volume-title":"\"The concept of tensorization\"","author":"Debals","year":"2017"},{"issue":"1","key":"2026033012250342600_ref032","doi-asserted-by":"crossref","first-page":"59","DOI":"10.1016\/0165-1684(95)00042-C","article-title":"\u201cAdaptive blind separation of independent sources: a deflation approach\u201d","volume":"45","author":"Delfosse","year":"1995","journal-title":"Signal processing"},{"issue":"3","key":"2026033012250342600_ref033","doi-asserted-by":"crossref","first-page":"211","DOI":"10.1007\/BF02288367","article-title":"\u201cThe approximation of one matrix by another of lower rank\u201d","volume":"1","author":"Eckart","year":"1936","journal-title":"Psychometrika"},{"key":"2026033012250342600_ref034","first-page":"534","author":"Feige","year":"2002"},{"issue":"397","key":"2026033012250342600_ref035","doi-asserted-by":"crossref","first-page":"249","DOI":"10.1080\/01621459.1987.10478427","article-title":"\u201cExploratory projection pursuit\u201d","volume":"82","author":"Friedman","year":"1987","journal-title":"Journal of the American statistical association"},{"key":"2026033012250342600_ref036","first-page":"359","volume-title":"Foundations of Computer Science, 1996. Proceedings., 37th Annual Symposium on","author":"Frieze","year":"1996"},{"key":"2026033012250342600_ref037","first-page":"797","volume-title":"Conference on Learning Theory","author":"Ge","year":"2015"},{"key":"2026033012250342600_ref038","doi-asserted-by":"crossref","first-page":"761","DOI":"10.1145\/2746539.2746616","volume-title":"Proceedings of the forty-seventh annual ACM symposium on Theory of computing","author":"Ge","year":"2015"},{"key":"2026033012250342600_ref039","first-page":"2741","volume-title":"International Conference on Machine Learning","author":"Ge","year":"2016"},{"key":"2026033012250342600_ref040","volume-title":"RANDOM","author":"Ge","year":"2015"},{"key":"2026033012250342600_ref041","volume-title":"Matrix Computations","author":"Golub","year":"1996"},{"key":"2026033012250342600_ref042","volume-title":"\u201cMatrix computations\u201d","author":"Golub","year":"1990"},{"issue":"12","key":"2026033012250342600_ref043","doi-asserted-by":"crossref","first-page":"613","DOI":"10.1016\/S0304-3975(00)00157-2","article-title":"\u201cLinear lower bound on degrees of Positivstellensatz calculus proofs for the parity\u201d","volume":"259","author":"Grigoriev","year":"2001","journal-title":"Theoretical Computer Science"},{"key":"2026033012250342600_ref044","volume-title":"\u201cFoundations of the PARAFAC procedure: models and conditions for an \u201cexplanatory\u2019 multimodal factor analysis\u201d","author":"Harshman","year":"1970"},{"issue":"1","key":"2026033012250342600_ref045","doi-asserted-by":"crossref","first-page":"39","DOI":"10.1016\/0167-9473(94)90132-5","article-title":"\u201cPARAFAC: Parallel factor analysis\u201d","volume":"18","author":"Harshman","year":"1994","journal-title":"Computational Statistics & Data Analysis"},{"issue":"6","key":"2026033012250342600_ref046","doi-asserted-by":"crossref","first-page":"45","DOI":"10.1145\/2512329","volume":"60","author":"Hillar","year":"2013","journal-title":"Journal of the ACM (JACM)"},{"issue":"1-4","key":"2026033012250342600_ref047","doi-asserted-by":"crossref","first-page":"164","DOI":"10.1002\/sapm192761164","article-title":"\u201cThe expression of a tensor or a polyadic as a sum of products\u201d","volume":"6","author":"Hitchcock","year":"1927","journal-title":"Journal of Mathematics and Physics"},{"key":"2026033012250342600_ref048","first-page":"289","volume-title":"Proceedings of the Fifteenth conference on Uncertainty in artificial intel ligence","author":"Hofmann","year":"1999"},{"key":"2026033012250342600_ref049","first-page":"956","volume-title":"Conference on Learning Theory","author":"Hopkins","year":"2015"},{"key":"2026033012250342600_ref050","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9781139020411","volume-title":"Matrix analysis","author":"Horn","year":"2012."},{"issue":"6","key":"2026033012250342600_ref051","doi-asserted-by":"crossref","first-page":"417","DOI":"10.1037\/h0071325","article-title":"\u201cAnalysis of a complex of statistical variables into principal components\u201d","volume":"24","author":"Hotelling","year":"1933.","journal-title":"Journal of educational psychology"},{"key":"2026033012250342600_ref052","doi-asserted-by":"crossref","first-page":"162","DOI":"10.1007\/978-1-4612-4380-9_14","volume-title":"Break-throughs in statistics","author":"Hotelling","year":"1992"},{"key":"2026033012250342600_ref053","volume-title":"Fourth Innovations in Theoretical Computer Science","author":"Hsu","year":"2013"},{"key":"2026033012250342600_ref054","volume-title":"Advances in Neural Information Processing Systems 25","author":"Hsu","year":"2012"},{"issue":"3","key":"2026033012250342600_ref055","doi-asserted-by":"crossref","first-page":"90","DOI":"10.1109\/MCSE.2007.55","article-title":"\u201cMatplotlib: A 2D Graphics Environment\u201d","volume":"9","author":"Hunter","year":"2007","journal-title":"Computing in Science Engineering"},{"issue":"4-5","key":"2026033012250342600_ref056","doi-asserted-by":"crossref","first-page":"411","DOI":"10.1016\/S0893-6080(00)00026-5","article-title":"\u201cIndependent component analysis: algorithms and applications\u201d","volume":"13","author":"Hyvarinen","year":"2000","journal-title":"Neural Networks"},{"key":"2026033012250342600_ref057","first-page":"695","volume-title":"Journal of Machine Learning Research","author":"Hyvarinen","year":"2005"},{"key":"2026033012250342600_ref058","volume-title":"\u201cScore Function Features for Discriminative Learning: Matrix and Tensor Frameworks\u201d","author":"Janzamin","year":"2014"},{"key":"2026033012250342600_ref059","volume-title":"\u201cBeating the perils of non-convexity: Guaranteed training of neural networks using tensor methods\u201d","author":"Janzamin","year":"2015"},{"key":"2026033012250342600_ref060","first-page":"315","volume-title":"Advances in neural information processing systems","author":"Johnson","year":"2013"},{"key":"2026033012250342600_ref061","volume-title":"\u201cSciPy: Open source scientific tools for Python\u201d","author":"Jones","year":"2001"},{"key":"2026033012250342600_ref062","volume-title":"\"Compression of Deep Convolutional Neural Networks for Fast and Low Power Mobile Applications\u201d","author":"Kim","year":"2016"},{"issue":"6","key":"2026033012250342600_ref063","doi-asserted-by":"crossref","first-page":"741","DOI":"10.1175\/1520-0450(1999)038<0741:TFODWT>2.0.CO;2","article-title":"\u201cThe feasibility of data whitening to improve performance of weather radar\u201d","volume":"38","author":"Koivunen","year":"1999","journal-title":"Journal of Applied Meteorology"},{"key":"2026033012250342600_ref064","volume-title":"\u201cStochastically Rank-Regularized Tensor Regression Networks\u201d","author":"Kolbeinsson","year":"2019"},{"issue":"4","key":"2026033012250342600_ref065","doi-asserted-by":"crossref","first-page":"1095","DOI":"10.1137\/100801482","article-title":"\u201cShifted Power Method for Computing Tensor Eigenpairs\u201d","volume":"32","author":"Kolda","year":"2011","journal-title":"SIAM Journal on Matrix Analysis and Applications"},{"issue":"3","key":"2026033012250342600_ref066","doi-asserted-by":"crossref","first-page":"455","DOI":"10.1137\/07070111X","article-title":"\u201cTensor decompositions and applications\u201d","volume":"51","author":"Kolda","year":"2009","journal-title":"SIAM review"},{"key":"2026033012250342600_ref067","first-page":"7822","author":"Kossaifi","year":"2019"},{"key":"2026033012250342600_ref068","doi-asserted-by":"crossref","first-page":"1940","DOI":"10.1109\/CVPRW.2017.243","volume-title":"Computer Vision and Pattern Recognition Workshops (CVPRW), 2017 IEEE Conference on","author":"Kossaifi","year":"2017"},{"key":"2026033012250342600_ref069","volume-title":"\u201cTensor Regression Networks\u201d","author":"Kossaifi","year":"2018"},{"issue":"26","key":"2026033012250342600_ref070","first-page":"1","article-title":"\u201cTensorLy: Tensor Learning in Python\u201d","volume":"20","author":"Kossaifi","year":"2019","journal-title":"Journal of Machine Learning Research"},{"issue":"3","key":"2026033012250342600_ref071","first-page":"281","volume":"41","author":"Kruskal","year":"1976","journal-title":"\u201cMore factors than subjects, tests and treatments: an indeterminacy theorem for canonical decomposition and individual differences scaling\u201d"},{"issue":"2","key":"2026033012250342600_ref072","doi-asserted-by":"crossref","first-page":"95","DOI":"10.1016\/0024-3795(77)90069-6","article-title":"\u201cThree-Way Arrays: Rank And Uniqueness Of TrilinEar Decompositions, With Application To Arithmetic Complexity And Statistics\u201d","volume":"18","author":"Kruskal","year":"1977","journal-title":"Linear algebra and its applications"},{"issue":"5","key":"2026033012250342600_ref073","doi-asserted-by":"crossref","first-page":"1273","DOI":"10.1090\/S0002-9939-04-07800-1","article-title":"\u201cSome estimates of norms of random matrices\u201d","volume":"133","author":"Lata\u0142a","year":"2005","journal-title":"Proceedings of the American Mathematical Society"},{"issue":"4","key":"2026033012250342600_ref074","doi-asserted-by":"crossref","first-page":"1324","DOI":"10.1137\/S0895479898346995","article-title":"\u201cOn the Best rank-1 and Rank-(R1, R2, ..., RN) Approximation and Applications of Higher-Order Tensors\u201d","volume":"21","author":"Lathauwer","year":"2000","journal-title":"SIAM J. Matrix Anal. Appl."},{"key":"2026033012250342600_ref075","volume-title":"\u201cSpeeding-up Convolutional Neural Networks Using Fine-tuned CP-Decomposition\u201d","author":"Lebedev","year":"2015"},{"issue":"4","key":"2026033012250342600_ref076","doi-asserted-by":"crossref","first-page":"1064","DOI":"10.1137\/0614071","article-title":"\u201cA decomposition for three-way arrays\u201d","volume":"14","author":"Leurgans","year":"1993","journal-title":"SIAM Journal on Matrix Analysis and Applications"},{"key":"2026033012250342600_ref077","first-page":"129","article-title":"\u201cSingular values and eigenvalues of tensors: a variational approach\u201d","volume":"1","author":"Lim","year":"2005","journal-title":"Proceedings of the IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP\u201905)"},{"key":"2026033012250342600_ref078","doi-asserted-by":"crossref","first-page":"438","DOI":"10.1109\/FOCS.2016.54","volume-title":"Foundations of Computer Science (FOCS), 2016 IEEE 57th Annual Symposium on","author":"Ma","year":"2016"},{"key":"2026033012250342600_ref079","first-page":"281","volume":"Vol. 1","author":"MacQueen","year":"1967","journal-title":"Proceedings of the fifth Berkeley Symposium on Mathematical Statistics and Probability"},{"issue":"35","key":"2026033012250342600_ref080","doi-asserted-by":"crossref","first-page":"482","DOI":"10.1109\/10.2119","article-title":"\u201cTopographic components model for event-related potentials and some biophysical considerations\u201d","volume":"6","author":"Mocks","year":"1988","journal-title":"IEEE transactions on biomedical engineering"},{"key":"2026033012250342600_ref081","first-page":"394","volume-title":"\u201cOn the reciprocal of the general algebraic matrix\u201d","author":"Moore","year":"1920"},{"issue":"2","key":"2026033012250342600_ref082","doi-asserted-by":"crossref","first-page":"583","DOI":"10.1214\/105051606000000024","article-title":"\u201cLearning Nonsingular Phylogenies and Hidden Markov Models\u201d","volume":"16","author":"Mossel","year":"2006","journal-title":"Annals of Applied Probability"},{"key":"2026033012250342600_ref083","volume-title":"\u201cTensor sparsification via a bound on the spectral norm of random tensors\u201d","author":"Nguyen","year":"2010"},{"key":"2026033012250342600_ref084","doi-asserted-by":"crossref","DOI":"10.1007\/b98874","volume-title":"Numerical Optimization","author":"Nocedal","year":"1999"},{"key":"2026033012250342600_ref085","first-page":"442","volume-title":"Proceedings of the 28th International Conference on Neural Information Processing Systems","author":"Novikov","year":"2015"},{"issue":"2","key":"2026033012250342600_ref086","doi-asserted-by":"crossref","first-page":"217","DOI":"10.1006\/jcss.2000.1711","article-title":"\u201cLatent semantic indexing: A probabilistic analysis\u201d","volume":"61","author":"Papadimitriou","year":"2000","journal-title":"Journal of Computer and System Sciences"},{"issue":"2","key":"2026033012250342600_ref087","first-page":"16","article-title":"\u201cTensors for data mining and data fusion: Models, applications, and scalable algorithms\u201d","volume":"8","author":"Papalexakis","year":"2017","journal-title":"ACM Transactions on Intel ligent Systems and Technology (TIST)"},{"key":"2026033012250342600_ref088","volume-title":"NIPS Autodiff Workshop","author":"Paszke","year":"2017"},{"key":"2026033012250342600_ref089","doi-asserted-by":"crossref","first-page":"71","DOI":"10.1098\/rsta.1894.0003","article-title":"\u201cContributions to the mathematical theory of evolution","volume":"185","author":"Pearson","year":"1894","journal-title":"Philosophical Transactions of the Royal Society of London. A"},{"issue":"11","key":"2026033012250342600_ref090","doi-asserted-by":"crossref","first-page":"559","DOI":"10.1080\/14786440109462720","article-title":"\u201cLIII. On lines and planes of closest fit to systems of points in space\u201d","volume":"2","author":"Pearson","year":"1901","journal-title":"The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science"},{"key":"2026033012250342600_ref091","first-page":"2825","article-title":"\u201cScikit-learn: Machine Learning in Python\u201d","volume":"12","author":"Pedregosa","year":"2011","journal-title":"Journal of Machine Learning Research"},{"issue":"No. 3","key":"2026033012250342600_ref092","doi-asserted-by":"crossref","first-page":"406","DOI":"10.1017\/S0305004100030401","article-title":"\u201cA generalized inverse for matrices\u201d","volume":"Vol. 51.","author":"Penrose","year":"1955","journal-title":"Mathematical proceedings of the Cambridge philosophical society"},{"key":"2026033012250342600_ref093","author":"Potechin","year":"2017","journal-title":"\u201cExact tensor completion with sum-of-squares\u201d"},{"issue":"6","key":"2026033012250342600_ref094","doi-asserted-by":"crossref","first-page":"1302","DOI":"10.1016\/j.jsc.2005.05.007","article-title":"\u201cEigenvalues of a real supersymmetric tensor\u201d","volume":"40","author":"Qi","year":"2005","journal-title":"Journal of Symbolic Computation"},{"key":"2026033012250342600_ref095","volume-title":"\u201cIntroduction to tensor decompositions and their applications in machine learning\u201d","author":"Rabanser","year":"2017"},{"issue":"6","key":"2026033012250342600_ref096","doi-asserted-by":"crossref","first-page":"40","DOI":"10.1145\/2535928","article-title":"\u201cTensor-rank and lower bounds for arithmetic formulas\u201d","volume":"60","author":"Raz","year":"2013","journal-title":"Journal of the ACM (JACM)"},{"key":"2026033012250342600_ref097","first-page":"2897","volume-title":"Advances in Neural Information Processing Systems","author":"Richard","year":"2014"},{"key":"2026033012250342600_ref098","first-page":"593","volume-title":"Foundations of Computer Science, 2008. FOCS\u201908. IEEE 49th Annual IEEE Symposium on","author":"Schoenebeck","year":"2008"},{"key":"2026033012250342600_ref099","first-page":"1223","volume-title":"Artificial Intel ligence and Statistics","author":"Sedghi","year":"2016"},{"issue":"Feb","key":"2026033012250342600_ref100","first-page":"567","article-title":"\u201cStochastic dual coordinate ascent methods for regularized loss minimization\u201d","volume":"14","author":"Shalev-Shwartz","year":"2013","journal-title":"Journal of Machine Learning Research"},{"key":"2026033012250342600_ref101","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2001.990454","article-title":"\u201cLinear image coding for regression and classification using the tensor-rank principle\u201d","volume":"Vol. 1","author":"Shashua","year":"2001","journal-title":"Computer Vision and Pattern Recognition, 2001. CVPR 2001. Proceedings of the 2001 IEEE Computer Society Conference on"},{"issue":"4","key":"2026033012250342600_ref102","doi-asserted-by":"crossref","first-page":"241","DOI":"10.1055\/s-0038-1634846","article-title":"\u201cProbabilistic diagnosis using a reformulation of the INTERNIST-1\/QMR knowledge base\u201d","volume":"30","author":"Shwe","year":"1991","journal-title":"Methods of information in Medicine"},{"issue":"3","key":"2026033012250342600_ref103","doi-asserted-by":"crossref","first-page":"229","DOI":"10.1002\/1099-128X(200005\/06)14:3<229::AID-CEM587>3.0.CO;2-N","article-title":"\u201cOn the uniqueness of multilinear decomposition of N-way arrays\u201d","volume":"14","author":"Sidiropoulos","year":"2000","journal-title":"Journal of Chemometrics"},{"issue":"8","key":"2026033012250342600_ref104","doi-asserted-by":"crossref","first-page":"2377","DOI":"10.1109\/78.852018","article-title":"\u201cParallel factor analysis in sensor array processing\u201d","volume":"48","author":"Sidiropoulos","year":"2000","journal-title":"Signal Processing, IEEE Transactions on"},{"issue":"13","key":"2026033012250342600_ref105","doi-asserted-by":"crossref","first-page":"3551","DOI":"10.1109\/TSP.2017.2690524","article-title":"\u201cTensor decomposition for signal processing and machine learning\u201d","volume":"65","author":"Sidiropoulos","year":"2017","journal-title":"IEEE Transactions on Signal Processing"},{"issue":"2","key":"2026033012250342600_ref106","doi-asserted-by":"crossref","first-page":"201","DOI":"10.2307\/1412107","article-title":"\u201cGeneral Intelligence,\u201d Objectively Determined and Measured\u201d","volume":"15","author":"Spearman","year":"1904","journal-title":"The American Journal of Psychology"},{"key":"2026033012250342600_ref107","volume-title":"\u201cDensity estimation in infinite dimensional exponential families\u201d","author":"Sriperumbudur","year":"2013"},{"key":"2026033012250342600_ref108","volume-title":"Matrix perturbation theory","author":"stewart","year":"1990"},{"key":"2026033012250342600_ref109","first-page":"1201","volume-title":"Proceedings of the 28th International Conference on Machine Learning (ICML-11)","author":"Swersky","year":"2011"},{"key":"2026033012250342600_ref110","volume-title":"\u201cConvolutional neural networks with low-rank regularization\u201d","author":"Tai","year":"2016"},{"issue":"4","key":"2026033012250342600_ref111","doi-asserted-by":"crossref","first-page":"389","DOI":"10.1007\/s10208-011-9099-z","article-title":"\u201cUser-friendly tail bounds for sums of random matrices\u201d","volume":"12","author":"Tropp","year":"2012","journal-title":"Foundations of computational mathematics."},{"issue":"1-2","key":"2026033012250342600_ref112","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1561\/2200000001","article-title":"\u201cGraphical models, exponential families, and variational inference\u201d","volume":"1","author":"Wainwright","year":"2008","journal-title":"Foundations and TrendsR in Machine Learning"},{"issue":"2","key":"2026033012250342600_ref113","doi-asserted-by":"crossref","first-page":"22","DOI":"10.1109\/MCSE.2011.37","article-title":"\u201cThe NumPy Array: A Structure for Efficient Numerical Computation\u201d","volume":"13","author":"Colbert","year":"2011","journal-title":"Computing in Science Engineering"},{"key":"2026033012250342600_ref114","volume-title":"Advances in Neural Information Processing Systems","author":"Wang","year":"2016"},{"issue":"1","key":"2026033012250342600_ref115","doi-asserted-by":"crossref","first-page":"99","DOI":"10.1007\/BF01932678","article-title":"\u201cPerturbation bounds in connection with singular value decomposition\u201d","volume":"12","author":"Wedin","year":"1972","journal-title":"BIT Numerical Mathematics"},{"issue":"4","key":"2026033012250342600_ref116","doi-asserted-by":"crossref","first-page":"441","DOI":"10.1007\/BF01456804","article-title":"\u201cDas asymptotische Verteilungsgesetz der Eigenwerte linearer partieller Differentialgleichungen (mit einer Anwendung auf die Theorie der Hohlraumstrahlung)\u201d","volume":"71","author":"Weyl","year":"1912","journal-title":"Mathematische Annalen"},{"key":"2026033012250342600_ref117","doi-asserted-by":"crossref","first-page":"534","DOI":"10.1137\/S0895479899352045","article-title":"\u201cRank-one approximation to high order tensors\u201d","volume":"23","author":"Zhang","year":"2001","journal-title":"SIAM Journal on Matrix Analysis and Applications"}],"container-title":["Foundations and Trends\u00ae in Machine Learning"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.emerald.com\/ftmal\/article-pdf\/12\/5-6\/393\/11164198\/2200000057en.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/www.emerald.com\/ftmal\/article-pdf\/12\/5-6\/393\/11164198\/2200000057en.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T18:10:47Z","timestamp":1777486247000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.emerald.com\/ftmal\/article\/12\/5-6\/393\/1333217\/Spectral-Learning-on-Matrices-and-Tensors"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,11,28]]},"references-count":117,"journal-issue":{"issue":"5-6","published-print":{"date-parts":[[2019,11,28]]}},"URL":"https:\/\/doi.org\/10.1561\/2200000057","relation":{},"ISSN":["1935-8237","1935-8245"],"issn-type":[{"value":"1935-8237","type":"print"},{"value":"1935-8245","type":"electronic"}],"subject":[],"published":{"date-parts":[[2019,11,28]]}}}