Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2610.10027

Efficient Optimization of Tensor Rings with Low-Rank Environments

Matthieu Jeannin, Alessandro Chessari, Jan von Delft

Published Oct 7, 2026• 0 citations
arXivHTMLPDF

Abstract

Tensor-ring (TR) decompositions provide a natural representation of periodic systems but are difficult to optimize because the closed geometry prevents a global canonical form and leads to costly, ill-conditioned environments. Existing periodic Density Matrix Renormalization Group (DMRG) methods alleviate this difficulty by compressing long environments to a low-rank representation, reducing local operations to $\mathcal{O}(pχ^3)$, where $p$ is the retained environment rank. Here, we extend this approach to an efficient two-site Ring-DMRG algorithm and improve its numerical robustness through appropriate gauge transformations and a generalized Davidson solver. A central challenge in the two-site formulation is the truncation step, which must account for the surrounding environment. When this environment is sufficiently separable, a suitable change of frame reduces the truncation to an ordinary singular value decomposition. When it is not separable, we instead use an alternating optimization that retains the full environment. The computational advantage of tensor rings over tensor trains depends on the scaling of the required environment rank $p$ with bond dimension and system size. For critical systems, the divergent correlation length makes long-range observables remain sensitive to the periodic geometry as the thermodynamic limit is approached. In this regime, the environment rank required to reach a fixed accuracy in the observable does not grow with the system size. Ring-DMRG therefore retains the same cubic scaling with bond dimension as standard DMRG, but at a substantially smaller bond dimension. The low-rank environment construction also provides a systematic generalization of belief propagation (BP), with standard BP recovered in the rank-one limit.

DOI
10.48550/arxiv.2610.10027

Related papers

Zero-site density matrix renormalization group and the optimal low-rank correction

Y. Núñez-Fernández, Gonzalo Torroba

A zero-site density matrix renormalization algorithm (DMRG0) is proposed to minimize the energy of matrix product states (MPSs). Instead of the site tensors themselves, we propose to optimize sequentially the ``message'' tensors between neighbor sites, which contain the singular values of the bipartition. This leads to a local minimization step that is independent of the physical dimension of the site. Conceptually, it separates the optimization and decimation steps in DMRG. Furthermore, we introduce two global perturbations based on the optimal low-rank correction to the current state, which are used to avoid local minima. They are determined variationally as the MPSs closest to the one-step correction of the Lanczos or Jacobi-Davidson eigensolver. These perturbations mainly decrease the energy and are free of hand-tuned parameters. Compared to existing single-site enrichment proposals, our approach gives similar convergence ratios per sweep, while the computations are cheaper by construction. Our methods may be useful in systems with many physical degrees of freedom per lattice site. We test our approach on the periodic Heisenberg spin chain for various spins and on free electrons on the lattice.

arXiv:1908.10880•Physical review. B./Physical review. B•Feb 24, 2020•2 citations

A Provably Efficient Method for Tensor Ring Decomposition and Its Applications

Chen, Han, Sitan Chen, Anru R. Zhang

We present the first deterministic, finite-step algorithm for exact tensor ring (TR) decomposition. Our method leverages blockwise simultaneous diagonalization to recover TR cores from a limited number of tensor observations, under a dimension condition requiring each mode size to be at least quadratic in the TR rank and under a genericity assumption on the cores, thereby providing both algebraic insight and practical efficiency. We extend the approach to the symmetric TR setting, where parameter complexity is significantly reduced and applications arise naturally in physics-based modeling and exchangeable data analysis. To handle noisy observations, we develop a robust recovery scheme that couples our initialization with alternating least squares, achieving faster convergence and improved accuracy compared to classical methods. As applications, we obtain new algorithms for questions in other domains where tensor ring decomposition is a key primitive, namely matrix product state tomography in quantum information and provable learning of pushforward distributions in the foundations of machine learning. These contributions advance the algorithmic foundations of TR decomposition and open new opportunities for scalable tensor network computation.

arXiv:2512.01016•Nov 30, 2025•1 citation

Enabling large-depth simulation of noisy quantum circuits with positive tensor networks

Ambroise Müller, Thomas Ayral, Corentin Bertrand

Matrix product density operators (MPDOs) are tensor network representations of locally purified density matrices where each physical degree of freedom is associated to an environment degree of freedom. MPDOs have interesting properties for mixed state representations: guaranteed positivity by construction, efficient conservation of the trace and computation of local observables. However, they have been challenging to use for noisy quantum circuit simulation, as the application of noise increases the dimension of the environment Hilbert space, leading to an exponential growth of bond dimensions. MPDOs also lack a unique canonical form, due to the freedom in the choice of basis for the environment Hilbert space, which leads to a vast variation of bond dimensions. In this work, we present a systematic way to reduce the bond dimensions of MPDOs by disentangling the purified state. We optimize the basis for the environment Hilbert space by performing density matrix renormalization group (DMRG)-like sweeps of local 2-qubit basis optimization. Interestingly, we find that targeting only the disentanglement of the purified state leads to a reduction of the environment dimension. In other words, a compact MPDO representation requires a low-entanglement purified state. We apply our compression method to the emulation of noisy random quantum circuits. Our technique allows us to keep bounded bond dimensions, and thus bounded memory, contrary to previous works on MPDOs, while keeping reasonable truncation fidelities.

arXiv:2403.00152•Feb 29, 2024•9 citations

Randomized density matrix renormalization group algorithm for low-rank tensor train decomposition

Huilin Jiang, Zhongming Chen, Gaohang Yu

Tensor train decomposition is a powerful tool for processing high-dimensional data. Density matrix renormalization group (DMRG) is an alternating scheme for low-rank tensor train decomposition of large tensors. However, it may suffer from the curse of dimensionality due to the large scale of subproblems. In this paper, we proposed a novel randomized proximal DMRG algorithm for low-rank tensor train decomposition by using TensorSketch to alleviate the curse of dimensionality. Numerical experiments on synthetic and real-world data also demonstrate the effectiveness and efficiency of the proposed algorithm.

Statistics Optimization & Information Computing•May 14, 2024•1 citation

Hybrid stochastic method for the tensor renormalization group

Hiroshi Ohki, Masaaki Tomii, Erika Arai

We propose a hybrid stochastic method for the tensor renormalization group (TRG) approach. TRG is known as a powerful tool to study the many-body systems and quantum field theory on the lattice. It is based on a low-rank approximation of the tensor using the truncated singular value decomposition (SVD), whose computational cost increases as the bond dimension increases, so that efficient cost reduction techniques are highly demanded. We use noise vectors for the low-rank approximation with the truncated SVD, by which the truncation error is replaced with a statistical error due to noise, and an error estimation could be improved. We test this method in the classical Ising model and observe a better accuracy than TRG. We also discuss a cross contamination issue in a multiple use of the same noise vectors, and to remove this systematic error we consider position-dependent noise vectors.

arXiv:2110.11921•International Symposium on Lattice Field Theory (LATTICE)•May 16, 2022•1 citation

On algorithms for and computing with the tensor ring decomposition

Oscar Mickelin, Sertaç Karaman

Abstract Tensor decompositions such as the canonical format and the tensor train format have been widely utilized to reduce storage costs and operational complexities for high‐dimensional data, achieving linear scaling with the input dimension instead of exponential scaling. In this paper, we investigate even lower storage‐cost representations in the tensor ring format, which is an extension of the tensor train format with variable end‐ranks. Firstly, we introduce two algorithms for converting a tensor in full format to tensor ring format with low storage cost. Secondly, we detail a rounding operation for tensor rings and show how this requires new definitions of common linear algebra operations in the format to obtain storage‐cost savings. Lastly, we introduce algorithms for transforming the graph structure of graph‐based tensor formats, with orders of magnitude lower complexity than existing literature. The efficiency of all algorithms is demonstrated on a number of numerical examples, and in certain cases, we demonstrate significantly higher compression ratios when compared to previous approaches to using the tensor ring format.

arXiv:1807.02513•Numerical Linear Algebra with Applications•Feb 24, 2020•46 citations
Tensor-ring (TR) decompositions provide a natural representation of periodic systems but are difficult to optimize because the closed geometry prevents a global canonical form and leads to costly, ill-conditioned environments. Existing periodic Density Matrix Renormalization Group (DMRG) methods alleviate this difficulty by compressing long environments to a low-rank representation, reducing local operations to O(pχ3)\mathcal{O}(pχ^3)O(pχ3), where ppp is the retained environment rank. Here, we extend this approach to an efficient two-site Ring-DMRG algorithm and improve its numerical robustness through appropriate gauge transformations and a generalized Davidson solver. A central challenge in the two-site formulation is the truncation step, which must account for the surrounding environment. When this environment is sufficiently separable, a suitable change of frame reduces the truncation to an ordinary singular value decomposition. When it is not separable, we instead use an alternating optimization that retains the full environment. The computational advantage of tensor rings over tensor trains depends on the scaling of the required environment rank ppp with bond dimension and system size. For critical systems, the divergent correlation length makes long-range observables remain sensitive to the periodic geometry as the thermodynamic limit is approached. In this regime, the environment rank required to reach a fixed accuracy in the observable does not grow with the system size. Ring-DMRG therefore retains the same cubic scaling with bond dimension as standard DMRG, but at a substantially smaller bond dimension. The low-rank environment construction also provides a systematic generalization of belief propagation (BP), with standard BP recovered in the rank-one limit.
A zero-site density matrix renormalization algorithm (DMRG0) is proposed to minimize the energy of matrix product states (MPSs). Instead of the site tensors themselves, we propose to optimize sequentially the ``message'' tensors between neighbor sites, which contain the singular values of the bipartition. This leads to a local minimization step that is independent of the physical dimension of the site. Conceptually, it separates the optimization and decimation steps in DMRG. Furthermore, we introduce two global perturbations based on the optimal low-rank correction to the current state, which are used to avoid local minima. They are determined variationally as the MPSs closest to the one-step correction of the Lanczos or Jacobi-Davidson eigensolver. These perturbations mainly decrease the energy and are free of hand-tuned parameters. Compared to existing single-site enrichment proposals, our approach gives similar convergence ratios per sweep, while the computations are cheaper by construction. Our methods may be useful in systems with many physical degrees of freedom per lattice site. We test our approach on the periodic Heisenberg spin chain for various spins and on free electrons on the lattice.
We present the first deterministic, finite-step algorithm for exact tensor ring (TR) decomposition. Our method leverages blockwise simultaneous diagonalization to recover TR cores from a limited number of tensor observations, under a dimension condition requiring each mode size to be at least quadratic in the TR rank and under a genericity assumption on the cores, thereby providing both algebraic insight and practical efficiency. We extend the approach to the symmetric TR setting, where parameter complexity is significantly reduced and applications arise naturally in physics-based modeling and exchangeable data analysis. To handle noisy observations, we develop a robust recovery scheme that couples our initialization with alternating least squares, achieving faster convergence and improved accuracy compared to classical methods. As applications, we obtain new algorithms for questions in other domains where tensor ring decomposition is a key primitive, namely matrix product state tomography in quantum information and provable learning of pushforward distributions in the foundations of machine learning. These contributions advance the algorithmic foundations of TR decomposition and open new opportunities for scalable tensor network computation.
Matrix product density operators (MPDOs) are tensor network representations of locally purified density matrices where each physical degree of freedom is associated to an environment degree of freedom. MPDOs have interesting properties for mixed state representations: guaranteed positivity by construction, efficient conservation of the trace and computation of local observables. However, they have been challenging to use for noisy quantum circuit simulation, as the application of noise increases the dimension of the environment Hilbert space, leading to an exponential growth of bond dimensions. MPDOs also lack a unique canonical form, due to the freedom in the choice of basis for the environment Hilbert space, which leads to a vast variation of bond dimensions. In this work, we present a systematic way to reduce the bond dimensions of MPDOs by disentangling the purified state. We optimize the basis for the environment Hilbert space by performing density matrix renormalization group (DMRG)-like sweeps of local 2-qubit basis optimization. Interestingly, we find that targeting only the disentanglement of the purified state leads to a reduction of the environment dimension. In other words, a compact MPDO representation requires a low-entanglement purified state. We apply our compression method to the emulation of noisy random quantum circuits. Our technique allows us to keep bounded bond dimensions, and thus bounded memory, contrary to previous works on MPDOs, while keeping reasonable truncation fidelities.
Tensor train decomposition is a powerful tool for processing high-dimensional data. Density matrix renormalization group (DMRG) is an alternating scheme for low-rank tensor train decomposition of large tensors. However, it may suffer from the curse of dimensionality due to the large scale of subproblems. In this paper, we proposed a novel randomized proximal DMRG algorithm for low-rank tensor train decomposition by using TensorSketch to alleviate the curse of dimensionality. Numerical experiments on synthetic and real-world data also demonstrate the effectiveness and efficiency of the proposed algorithm.
We propose a hybrid stochastic method for the tensor renormalization group (TRG) approach. TRG is known as a powerful tool to study the many-body systems and quantum field theory on the lattice. It is based on a low-rank approximation of the tensor using the truncated singular value decomposition (SVD), whose computational cost increases as the bond dimension increases, so that efficient cost reduction techniques are highly demanded. We use noise vectors for the low-rank approximation with the truncated SVD, by which the truncation error is replaced with a statistical error due to noise, and an error estimation could be improved. We test this method in the classical Ising model and observe a better accuracy than TRG. We also discuss a cross contamination issue in a multiple use of the same noise vectors, and to remove this systematic error we consider position-dependent noise vectors.
Abstract Tensor decompositions such as the canonical format and the tensor train format have been widely utilized to reduce storage costs and operational complexities for high‐dimensional data, achieving linear scaling with the input dimension instead of exponential scaling. In this paper, we investigate even lower storage‐cost representations in the tensor ring format, which is an extension of the tensor train format with variable end‐ranks. Firstly, we introduce two algorithms for converting a tensor in full format to tensor ring format with low storage cost. Secondly, we detail a rounding operation for tensor rings and show how this requires new definitions of common linear algebra operations in the format to obtain storage‐cost savings. Lastly, we introduce algorithms for transforming the graph structure of graph‐based tensor formats, with orders of magnitude lower complexity than existing literature. The efficiency of all algorithms is demonstrated on a number of numerical examples, and in certain cases, we demonstrate significantly higher compression ratios when compared to previous approaches to using the tensor ring format.