66
views
0
recommends
+1 Recommend
0 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      The trouble with tensor ring decompositions

      Preprint

      Read this article at

      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          The tensor train decomposition decomposes a tensor into a "train" of 3-way tensors that are interconnected through the summation of auxiliary indices. The decomposition is stable, has a well-defined notion of rank and enables the user to perform various linear algebra operations on vectors and matrices of exponential size in a computationally efficient manner. The tensor ring decomposition replaces the train by a ring through the introduction of one additional auxiliary variable. This article discusses a major issue with the tensor ring decomposition: its inability to compute an exact minimal-rank decomposition from a decomposition with sub-optimal ranks. Both the contraction operation and Hadamard product are motivated from applications and it is shown through simple examples how the tensor ring-rounding procedure fails to retrieve minimal-rank decompositions with these operations. These observations, together with the already known issue of not being able to find a best low-rank tensor ring approximation to a given tensor indicate that the applicability of tensor rings is severely limited.

          Related collections

          Most cited references18

          • Record: found
          • Abstract: found
          • Article: found
          Is Open Access

          The density-matrix renormalization group in the age of matrix product states

          The density-matrix renormalization group method (DMRG) has established itself over the last decade as the leading method for the simulation of the statics and dynamics of one-dimensional strongly correlated quantum lattice systems. In the further development of the method, the realization that DMRG operates on a highly interesting class of quantum states, so-called matrix product states (MPS), has allowed a much deeper understanding of the inner structure of the DMRG method, its further potential and its limitations. In this paper, I want to give a detailed exposition of current DMRG thinking in the MPS language in order to make the advisable implementation of the family of DMRG algorithms in exclusively MPS terms transparent. I then move on to discuss some directions of potentially fruitful further algorithmic development: while DMRG is a very mature method by now, I still see potential for further improvements, as exemplified by a number of recently introduced algorithms.
            Bookmark
            • Record: found
            • Abstract: not found
            • Article: not found

            Tensor-Train Decomposition

              Bookmark
              • Record: found
              • Abstract: not found
              • Article: not found

              TT-cross approximation for multidimensional arrays

                Bookmark

                Author and article information

                Journal
                09 November 2018
                Article
                1811.03813
                471e9e60-32c0-43d2-85b5-3ed24b2d7b99

                http://arxiv.org/licenses/nonexclusive-distrib/1.0/

                History
                Custom metadata
                cs.NA

                Numerical & Computational mathematics
                Numerical & Computational mathematics

                Comments

                Comment on this article