In this paper, we propose new learning algorithms for approximating high-dimensional functions using tree tensor networks in a least-squares setting. Given a dimension tree or architecture of the tensor network, we provide an algorithm that generates a sequence of nested tensor subspaces based on a generalization of principal component analysis for multivariate functions. An optimal least-squares method is used for computing projections onto the generated tensor subspaces, using samples generated from a distribution depending on the previously generated subspaces. We provide an error bound in expectation for the obtained approximation. Practical strategies are proposed for adapting the feature spaces and ranks to achieve a prescribed error. Also, we propose an algorithm that progressively constructs the dimension tree by suitable pairings of variables, that allows to further reduce the number of samples necessary to reach that error. Numerical examples illustrate the performance of the proposed algorithms and show that stable approximations are obtained with a number of samples close to the number of free parameters of the estimated tensor networks.
Approximation by tree tensor networks in high dimensions: Sobolev and compositional functions
Johannes Gutenberg-Universität Mainz(约翰内斯·古腾堡美因茨大学);Centrale Nantes, Nantes Université, Laboratoire de Mathématiques Jean Leray, CNRS UMR 6629(南特中央理工学院,南特大学,让·勒雷数学实验室,法国国家科学研究中心);Technische Universität Berlin(柏林工业大学)
Geometry of tree-based tensor formats in tensor Banach spaces
在论文“On the Dirac-Frenkel Variational Principle on Tensor巴拿赫空间”中,我们给出了张量巴拿赫空间中具有固定多线性秩的Tucker形式的张量流形的几何描述,这使得Dirac-Frenkel变分原理在拓扑张量空间的框架下得到了推广.本文的目的是将这些结果推广到更一般的张量格式。更确切地说,我们提供了一个新的几何描述的流形的张量在树为基础的(或层次)的格式,也被称为树张量网络,这是交叉的流形张量在塔克格式与不同的分区集的维度。本文提出的张量的树型几何描述与张量流形的Tucker几何描述是相容的。
Geometric Structures in Tensor Representations (Final Release)
Universidad CEU Cardenal Herrera(CEU卡迪纳尔埃雷拉大学);Max-Planck-Institut Mathematik in den Naturwissenschaften(马克斯·普朗克自然科学数学研究所);Ecole Centrale Nantes(南特中央理工学院)
Institute of Mathematics of the Polish Academy of Sciences(波兰科学院数学研究所);Faculty of Mathematics, Computer Science and Mechanics University of Warsaw(华沙大学数学、计算机和力学学院);Simons Institute for the Theory of Computing(西蒙斯计算理论研究所);UC Berkeley(加州大学伯克利分校)
Learning high-dimensional probability distributions using tree tensor networks
Centrale Nantes(南特中央理工学院);Laboratoire de Mathématiques Jean Leray(让·勒雷数学实验室);CNRS(法国国家科学研究中心);Naval Group(海军集团);Airbus Central Research & Technology(空客中央研究与技术中心)