Riemannian optimization of isometric tensor networks

Abstract

Several tensor networks are built of isometric tensors, i.e. tensors satisfying $W^{†}W=\mathrm{I}$. 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.

arXiv comment18 pages + appendices, 3 figures; v3 submission to SciPost; v4 expand preconditioning discussion and add polish, resubmit to SciPost

References in corpus (38)

• A variational eigenvalue solver on a quantum processor
• The density-matrix renormalization group in the age of matrix product states
• Matrix Product States, Projected Entangled Pair States, and variational renormalization group methods for quantum spin systems
• Efficient simulation of one-dimensional quantum many-body systems
• A class of quantum many-body states that can be efficiently simulated
• Time-dependent variational principle for quantum lattices
• Efficient variational quantum simulator incorporating active error minimisation
• Tensor renormalization group approach to 2D classical lattice models
• Classical simulation of quantum many-body systems with a tree tensor network
• Renormalization algorithms for Quantum-Many Body Systems in two and higher dimensions
• The iTEBD algorithm beyond unitary evolution
• Quantum Natural Gradient
• Theory of variational quantum simulation
• Tensor Network Renormalization
• Algorithms for entanglement renormalization
• Continuous Matrix Product States for Quantum Fields
• Variational optimization algorithms for uniform matrix product states
• Spin-waves in triangular lattice antiferromagnet: decays, spectrum renormalization, and singularities
• Variational optimization with infinite projected entangled-pair states
• Post-Matrix Product State Methods: To tangent space and beyond
• Era of Big Data Processing: A New Approach via Tensor Networks and Tensor Decompositions
• Gradient methods for variational optimization of projected entangled-pair states
• Tangent-space methods for uniform matrix product states
• Excitation spectra of the spin-1/2 triangular-lattice Heisenberg antiferromagnet
• Variational matrix product ansatz for dispersion relations
• Isometric Tensor Network States in Two Dimensions
• Dynamical structure factor of triangular-lattice antiferromagnet
• Parallel Quantum Simulation of Large Systems on Small Quantum Computers
• Supervised Learning with Quantum-Inspired Tensor Networks
• Geometry of variational methods: dynamics of closed quantum systems
• Matrix product states and variational methods applied to critical quantum field theory
• Geometry of Matrix Product States: metric, parallel transport and curvature
• Unifying variational methods for simulating quantum many-body systems
• Isometric Tensor Network representation of string-net liquids
• Continuous Matrix Product States for Quantum Fields: an Energy Minimization Algorithm
• Multigrid Algorithms for Tensor Network States
• Quantum magnetic properties of the spin-1/2 triangular-lattice antiferromagnet Ba${}_2$La${}_2$CoTe${}_2$O${}_{12}$
• Variational optimization of continuous matrix product states

Cited by in corpus (45)

• The ITensor Software Library for Tensor Network Calculations
• Tensor Network Algorithms: a Route Map
• Preparation of matrix product states with log-depth quantum circuits
• Classical Shadow Tomography with Locally Scrambled Quantum Dynamics
• Scalable and Flexible Classical Shadow Tomography with Tensor Networks
• Riemannian geometry and automatic differentiation for optimization problems of quantum physics and quantum technologies
• Tensor networks for quantum machine learning
• Direct sampling of projected entangled-pair states
• Efficient Simulation of Dynamics in Two-Dimensional Quantum Spin Systems with Isometric Tensor Networks
• Scaling of neural-network quantum states for time evolution
• Optimizing quantum circuits with Riemannian gradient flow
• Tree tensor network classifiers for machine learning: from quantum-inspired to quantum-assisted
• Variational methods for contracting projected entangled-pair states
• QGOpt: Riemannian optimization for quantum technologies
• Approximate encoding of quantum states using shallow circuits
• Isometric tensor network optimization for extensive Hamiltonians is free of barren plateaus
• Tensor networks for quantum computing
• Typical Correlation Length of Sequentially Generated Tensor Network States
• Quantum-classical eigensolver using multiscale entanglement renormalization
• Differentiable Programming of Isometric Tensor Networks
• Riemannian quantum circuit optimization for Hamiltonian simulation
• A tensor network discriminator architecture for classification of quantum data on quantum computers
• Rényi free energy and variational approximations to thermal states
• Efficient Quantum Circuit Compilation for Near-Term Quantum Advantage
• U(1)-symmetric Gaussian fermionic projected entangled paired states and their Gutzwiller projection
• Absence of barren plateaus and scaling of gradients in the energy optimization of isometric tensor network states
• Fermionic Isometric Tensor Network States in Two Dimensions
• Convergence and Quantum Advantage of Trotterized MERA for Strongly-Correlated Systems
• Double-bracket quantum algorithms for quantum imaginary-time evolution
• Riemannian quantum circuit optimization based on matrix product operators
• The Haldane gap in the SU(3) [3 0 0] Heisenberg chain
• Variational quantum eigensolver with embedded entanglement using a tensor-network ansatz
• Learning topological states from randomized measurements using variational tensor network tomography
• Controlling quantum many-body systems using reduced-order modelling
• Simulating quantum circuits using the multi-scale entanglement renormalization ansatz
• Two Dimensional Isometric Tensor Networks on an Infinite Strip
• Automatic Structural Search of Tensor Network States including Entanglement Renormalization
• Typical Machine Learning Datasets as Low-Depth Quantum Circuits
• A Riemannian Approach to the Lindbladian Dynamics of a Locally Purified Tensor Network
• Accelerating two-dimensional tensor network optimization by preconditioning
• Scaling of contraction costs for entanglement renormalization algorithms including tensor Trotterization and variational Monte Carlo
• Diagonal Isometric Form for Tensor Network States in Two Dimensions
• Fast Tensor Disentangling Algorithm
• Kac-Moody symmetries in one-dimensional bosonic systems
• Locally Purified Maximally Mixed States At Scale: Entanglement Pruning and Symmetries