Abstract:Although tensor networks are powerful tools for simulating low-dimensional quantum physics, tensor network algorithms are very computationally costly in higher spatial dimensions. We introduce quantum gauge networks: a different kind of tensor network ansatz for which the computation cost of simulations does not explicitly increase for larger spatial dimensions. We take inspiration from the gauge picture of quantum dynamics, which consists of a local wavefunction for each patch of space, with neighboring patches related by unitary connections. A quantum gauge network (QGN) has a similar structure, except the Hilbert space dimensions of the local wavefunctions and connections are truncated. We describe how a QGN can be obtained from a generic wavefunction or matrix product state (MPS). All $2k$-point correlation functions of any wavefunction for $M$ many operators can be encoded exactly by a QGN with bond dimension $O(M^k)$. In comparison, for just $k=1$, an exponentially larger bond dimension of $2^{M/6}$ is generically required for an MPS of qubits. We provide a simple QGN algorithm for approximate simulations of quantum dynamics in any spatial dimension. The approximate dynamics can achieve exact energy conservation for time-independent Hamiltonians, and spatial symmetries can also be maintained exactly. We benchmark the algorithm by simulating the quantum quench of fermionic Hamiltonians in up to three spatial dimensions.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the problem of excessive computational cost in traditional tensor network algorithms when performing quantum physics simulations in high - dimensional spaces. Specifically: 1. **Background and Challenges**: - Although tensor network algorithms are very effective in low - dimensional quantum physics simulations, the computational cost in high - dimensional spaces rises sharply as the dimension increases, resulting in low efficiency in practical applications. - The computational complexity of traditional tensor network methods (such as PEPS) in three - dimensional space is usually $O(\chi^{12})$, where $\chi$ is the bond dimension, which makes high - dimensional simulations very time - consuming. 2. **Proposed New Method**: - The paper introduces a new tensor network representation method - Quantum Gauge Networks (QGN). The design of QGN is inspired by the gauge picture of quantum dynamics and describes quantum states through local wave functions and gauge connections. - The main advantage of QGN is that even in high - dimensional spaces, its computational cost does not increase significantly with the increase of the spatial dimension. Specifically, the computational time complexity of the QGN algorithm is $O(\chi^3)$, which is the same as the matrix product state (MPS) algorithm in one - dimensional space. 3. **Specific Objectives**: - **Improve Computational Efficiency**: By truncating the dimension of the Hilbert space, QGN can significantly reduce the computational cost while maintaining high precision. - **Accurately Encode Correlation Functions**: QGN can accurately encode the multi - point correlation functions of any wave function. In particular, for $2k$-point correlation functions, only $O(M^k)$ bond dimension is required, while the traditional MPS method may require an exponential bond dimension. - **Maintain Physical Properties**: The QGN algorithm can achieve energy conservation and can accurately maintain spatial symmetry. 4. **Application Scenarios**: - The paper verifies the effectiveness of the QGN algorithm in two - dimensional and three - dimensional spaces by simulating the quantum quenching process of the fermionic Hamiltonian. In summary, this paper aims to overcome the problem of excessive computational cost of traditional tensor network algorithms in high - dimensional spaces by introducing Quantum Gauge Networks (QGN) and providing a new method for efficiently simulating quantum physics systems in high - dimensional spaces.

Quantum Gauge Networks: A New Kind of Tensor Network

Quantum and classical spin network algorithms for $q$-deformed Kogut-Susskind gauge theories

Quantum tensor network algorithms for evaluation of spectral functions on quantum computers

Gauge-Fixing Quantum Density Operators At Scale

Gauge Invariant and Anyonic Symmetric Transformer and RNN Quantum States for Quantum Lattice Models

Three-dimensional isometric tensor networks

Quantum-Inspired Fluid Simulation of 2D Turbulence with GPU Acceleration

Finding the ground state of a lattice gauge theory with fermionic tensor networks: a $2+1d$ $\mathbb{Z}_2$ demonstration

Quantum Circuits For Two-Dimensional Isometric Tensor Networks

Exact holographic tensor networks for the Motzkin spin chain

Tensor Networks for Lattice Gauge Theories beyond one dimension: a Roadmap

The Gauge Picture of Quantum Dynamics

Quantum Graph Neural Networks

Gauged Gaussian PEPS -- A High Dimensional Tensor Network Formulation for Lattice Gauge Theories

Suppressing Gauge Drift in Quantum Simulations with Gauge Transformations

Spontaneously interacting qubits from Gauss-Bonnet

Gauged Gaussian projected entangled pair states: A high dimensional tensor network formulation for lattice gauge theories

Typical Correlation Length of Sequentially Generated Tensor Network States

Tensor network simulation of the quantum Kibble-Zurek quench from the Mott to the superfluid phase in the two-dimensional Bose-Hubbard model