{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T16:38:48Z","timestamp":1787330328689,"version":"build-2736575974"},"reference-count":40,"publisher":"Society for Industrial & Applied Mathematics (SIAM)","issue":"4","funder":[{"DOI":"10.13039\/501100003443","name":"Ministry of Education and Science of the Russian Federation","doi-asserted-by":"publisher","award":["075-10-2021-068"],"award-info":[{"award-number":["075-10-2021-068"]}],"id":[{"id":"10.13039\/501100003443","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["SIAM J. Sci. Comput."],"published-print":{"date-parts":[[2023,8,31]]},"abstract":"<jats:p>Abstract.<\/jats:p>\n                  <jats:p>Surrogate models can reduce computational costs for multivariable functions with an unknown internal structure (black boxes). In a discrete formulation, surrogate modeling is equivalent to restoring a multidimensional array (tensor) from a small part of its elements. The alternating least squares (ALS) algorithm in the tensor train (TT) format is a widely used approach to effectively solve this problem in the case of nonadaptive tensor recovery from a given training set (i.e., tensor completion problem). TT-ALS allows obtaining a low-parametric representation of the tensor, which is free from the curse of dimensionality and can be used for fast computation of the values at arbitrary tensor indices or efficient implementation of algebra operations with the black box (integration, etc.). However, to obtain high accuracy in the presence of restrictions on the size of the train data, a good choice of initial approximation is essential. In this work, we construct the ANOVA representation in the TT-format and use it as an initial approximation for the TT-ALS algorithm. The performed numerical computations for a number of multidimensional model problems, including the parametric partial differential equation, demonstrate a significant advantage of our approach for the commonly used random initial approximation. For all considered model problems we obtained an increase in accuracy by at least an order of magnitude with the same number of requests to the black box. The proposed approach is very general and can be applied in a wide class of real-world surrogate modeling and machine learning problems.<\/jats:p>","DOI":"10.1137\/22m1514088","type":"journal-article","created":{"date-parts":[[2023,8,11]],"date-time":"2023-08-11T04:01:21Z","timestamp":1691726481000},"page":"A2101-A2118","source":"Crossref","is-referenced-by-count":11,"title":["Black Box Approximation in the Tensor Train Format Initialized by ANOVA Decomposition"],"prefix":"10.1137","volume":"45","author":[{"given":"Andrei","family":"Chertkov","sequence":"first","affiliation":[{"name":"Skolkovo Institute of Science and Technology, Moscow, 121205, Russia."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gleb","family":"Ryzhakov","sequence":"additional","affiliation":[{"name":"Skolkovo Institute of Science and Technology, Moscow, 121205, Russia."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ivan","family":"Oseledets","sequence":"additional","affiliation":[{"name":"Skolkovo Institute of Science and Technology, Moscow, 121205, Russia, and AIRI, Moscow, Russia."}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"351","published-online":{"date-parts":[[2023,8,11]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1080\/00949659708811825"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1137\/S0036142902418680"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1137\/140960980"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijar.2021.11.007"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1016\/j.ress.2018.11.007"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.4324\/9780203763933"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2021.08.034"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.3389\/frai.2021.668215"},{"key":"ref9","unstructured":"A. Chertkov , \nI. Oseledets , and \nM. Rakhuba , Robust Discretization in Quantized Tensor Train Format for Elliptic Problems in Two Dimensions, preprint, https:\/\/arxiv.org\/abs\/1612.01166, 2016."},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1561\/2200000059"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1561\/2200000067"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1007\/s00791-011-0160-x"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.4236\/am.2012.330215"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1137\/17M1138881"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1137\/19M1244172"},{"key":"ref16","unstructured":"L. Grasedyck , \nM. Kluge , and \nS. Kr\u00e4mer , Alternating Directions Fitting (ADF) of Hierarchical Low Rank Tensors, preprint, DFG-Schwerpunktprogramm 1324, 2013."},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1137\/100818893"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1007\/s10107-022-01867-8"},{"key":"ref19","first-page":"150","volume":"4","author":"Jamil M.","year":"2013","journal-title":"J. Math. Model. Numer. Optim."},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1137\/19M1306518"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2020.2995061"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-23099-8"},{"key":"ref23","unstructured":"A. Nikitin , \nA. Chertkov , \nR. Ballester-Ripoll , \nI. Oseledets , and \nE. Frolov , Are Quantum Computers Practical Yet? A Case for Feature Selection in Recommender Systems using Tensor Networks, preprint, https:\/\/arxiv.org\/abs\/2205.04490, 2022."},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1137\/090752286"},{"key":"ref25","doi-asserted-by":"crossref","unstructured":"I. Oseledets , \nM. Rakhuba , and \nA. Chertkov , Black-box solver for multiscale modelling using the QTT format, in Proceedings of ECCOMAS, 2016.","DOI":"10.7712\/100016.2387.10906"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1137\/090748330"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2019.2956926"},{"key":"ref28","unstructured":"L. Richter , \nL. Sallandt , and \nN. N\u00fcsken , Solving high-dimensional parabolic PDEs using the tensor train format, in Proceedings of the International Conference on Machine Learning, PMLR, 2021, pp. 8998\u20139009."},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1137\/110857520"},{"key":"ref30","unstructured":"G. Ryzhakov  and \nI. Oseledets , Constructive TT-representation of the tensors given as index interaction functions with applications, in Proceedings of Eleventh International Conference on Learning Representations (ICLR-2023), 2023."},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.3389\/frai.2021.687176"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2021.3051503"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1016\/S0378-4754(00)00270-6"},{"key":"ref34","doi-asserted-by":"crossref","unstructured":"K. Sozykin , \nA. Chertkov , \nR. Schutski , \nA.H. Phan , \nA. Cichocki , and \nI. Oseledets , TTOpt: A maximum volume quantized tensor train-based optimization and its application to reinforcement learning, in Proceedings of Thirty-sixth Conference on Neural Information Processing Systems (NeurIPS-2022), 2022.","DOI":"10.52202\/068431-1889"},{"key":"ref35","unstructured":"C. Tobler , Low-Rank Tensor Methods for Linear Systems and Eigenvalue Problems, Ph.D. thesis, ETH Zurich, 2012."},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1137\/1.9781611975949"},{"key":"ref37","unstructured":"C. Vanaret , \nJ.B. Gotteland , \nN. Durand , and \nJ.M. Alliot , Certified Global Minima for a Benchmark of Difficult Optimization Problems, preprint, https:\/\/arxiv.org\/abs\/2003.09867, 2020."},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1007\/s10589-022-00428-1"},{"key":"ref39","unstructured":"V. Zankin , \nG. Ryzhakov , and \nI. Oseledets , Gradient Descent-Based D-Optimal Design for the Least-Squares Polynomial Approximation, preprint, https:\/\/arxiv.org\/abs\/1806.06631, 2018."},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2014.2369505"}],"container-title":["SIAM Journal on Scientific Computing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/epubs.siam.org\/doi\/pdf\/10.1137\/22M1514088","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T15:58:02Z","timestamp":1787327882000},"score":1,"resource":{"primary":{"URL":"https:\/\/epubs.siam.org\/doi\/10.1137\/22M1514088"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,8,11]]},"references-count":40,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2023,8,31]]}},"alternative-id":["10.1137\/22M1514088"],"URL":"https:\/\/doi.org\/10.1137\/22m1514088","relation":{},"ISSN":["1064-8275","1095-7197"],"issn-type":[{"value":"1064-8275","type":"print"},{"value":"1095-7197","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,8,11]]}}}