Riemannian optimization of isometric tensor networks

• URL: http://arxiv.org/abs/2007.03638v4
• Date: Wed, 13 Jan 2021 12:58:17 GMT
• Title: Riemannian optimization of isometric tensor networks
• Authors: Markus Hauru, Maarten Van Damme, and Jutho Haegeman
• Abstract summary: We show how gradient-based optimization methods can be used to optimize tensor networks of isometries to represent e.g. ground states of 1D quantum Hamiltonians. We apply these methods in the context of infinite MPS and MERA, and show benchmark results in which they outperform the best previously-known optimization methods.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Several tensor networks are built of isometric tensors, i.e. tensors satisfying $W^\dagger W = \mathrm{I}$. Prominent examples include matrix product states (MPS) in canonical form, the multiscale entanglement renormalization ansatz (MERA), and quantum circuits in general, such as those needed in state preparation and quantum variational eigensolvers. We show how gradient-based optimization methods on Riemannian manifolds can be used to optimize tensor networks of isometries to represent e.g. ground states of 1D quantum Hamiltonians. We discuss the geometry of Grassmann and Stiefel manifolds, the Riemannian manifolds of isometric tensors, and review how state-of-the-art optimization methods like nonlinear conjugate gradient and quasi-Newton algorithms can be implemented in this context. We apply these methods in the context of infinite MPS and MERA, and show benchmark results in which they outperform the best previously-known optimization methods, which are tailor-made for those specific variational classes. We also provide open-source implementations of our algorithms.

Related papers

• Quick design of feasible tensor networks for constrained combinatorial optimization [1.8775413720750924]
  In recent years, tensor networks have been applied to constrained optimization problems for practical applications. One approach is to construct tensor networks with nilpotent-matrix manipulation. The proposed method is expected to facilitate the discovery of feasible tensor networks for constrained optimization problems.
  arXiv  Detail & Related papers  (2024-09-03T08:36:23Z)
• FORML: A Riemannian Hessian-free Method for Meta-learning on Stiefel Manifolds [4.757859522106933]
  This paper introduces a Hessian-free approach that uses a first-order approximation of derivatives on the Stiefel manifold. Our method significantly reduces the computational load and memory footprint.
  arXiv  Detail & Related papers  (2024-02-28T10:57:30Z)
• Posterior Contraction Rates for Mat\'ern Gaussian Processes on Riemannian Manifolds [51.68005047958965]
  We show that intrinsic Gaussian processes can achieve better performance in practice. Our work shows that finer-grained analyses are needed to distinguish between different levels of data-efficiency.
  arXiv  Detail & Related papers  (2023-09-19T20:30:58Z)
• One-step replica symmetry breaking in the language of tensor networks [0.913755431537592]
  We develop an exact mapping between the one-step replica symmetry breaking cavity method and tensor networks. The two schemes come with complementary mathematical and numerical toolboxes that could be leveraged to improve the respective states of the art.
  arXiv  Detail & Related papers  (2023-06-26T18:42:51Z)
• GloptiNets: Scalable Non-Convex Optimization with Certificates [61.50835040805378]
  We present a novel approach to non-cube optimization with certificates, which handles smooth functions on the hypercube or on the torus. By exploiting the regularity of the target function intrinsic in the decay of its spectrum, we allow at the same time to obtain precise certificates and leverage the advanced and powerful neural networks.
  arXiv  Detail & Related papers  (2023-06-26T09:42:59Z)
• Adaptive Log-Euclidean Metrics for SPD Matrix Learning [73.12655932115881]
  We propose Adaptive Log-Euclidean Metrics (ALEMs), which extend the widely used Log-Euclidean Metric (LEM) The experimental and theoretical results demonstrate the merit of the proposed metrics in improving the performance of SPD neural networks.
  arXiv  Detail & Related papers  (2023-03-26T18:31:52Z)
• Differentiable Programming of Isometric Tensor Networks [0.0]
  Differentiable programming is a new programming paradigm which enables large scale optimization through automatic calculation of gradients also known as auto-differentiation. Here, we extend the differentiable programming to tensor networks with isometric constraints with applications to multiscale entanglement renormalization ansatz (MERA) and tensor network renormalization (TNR) We numerically tested our methods on 1D critical quantum Ising spin chain and 2D classical Ising model. We calculate the ground state energy for the 1D quantum model and internal energy for the classical model, and scaling dimensions of scaling operators and find they all agree
  arXiv  Detail & Related papers  (2021-10-08T05:29:41Z)
• Automatic differentiation for Riemannian optimization on low-rank matrix and tensor-train manifolds [71.94111815357064]
  In scientific computing and machine learning applications, matrices and more general multidimensional arrays (tensors) can often be approximated with the help of low-rank decompositions. One of the popular tools for finding the low-rank approximations is to use the Riemannian optimization.
  arXiv  Detail & Related papers  (2021-03-27T19:56:00Z)
• Efficient construction of tensor-network representations of many-body Gaussian states [59.94347858883343]
  We present a procedure to construct tensor-network representations of many-body Gaussian states efficiently and with a controllable error. These states include the ground and thermal states of bosonic and fermionic quadratic Hamiltonians, which are essential in the study of quantum many-body systems.
  arXiv  Detail & Related papers  (2020-08-12T11:30:23Z)
• Stochastic Flows and Geometric Optimization on the Orthogonal Group [52.50121190744979]
  We present a new class of geometrically-driven optimization algorithms on the orthogonal group $O(d)$. We show that our methods can be applied in various fields of machine learning including deep, convolutional and recurrent neural networks, reinforcement learning, flows and metric learning.
  arXiv  Detail & Related papers  (2020-03-30T15:37:50Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.