F. Fröwis,V. Nebendahl,W. Dür
Abstract:We consider the representation of operators in terms of tensor networks and their application to ground-state approximation and time evolution of systems with long-range interactions. We provide an explicit construction to represent an arbitrary many-body Hamilton operator in terms of a one-dimensional tensor network, i.e. as a matrix product operator. For pairwise interactions, we show that such a representation is always efficient and requires a tensor dimension growing only linearly with the number of particles. For systems obeying certain symmetries or restrictions we find optimal representations with minimal tensor dimension. We discuss the analytic and numerical approximation of operators in terms of low-dimensional tensor operators. We demonstrate applications for time evolution and ground-state approximation, in particular for long-range interaction with inhomogeneous couplings. The operator representations are also generalized to other geometries such as trees and 2D lattices, where we show how to obtain and use efficient tensor network representations respecting a given geometry.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is how to effectively represent operators of systems with long - range interactions, especially Hamiltonians, for ground - state approximation and time - evolution calculations. The author proposes a method of using one - dimensional tensor networks (i.e., matrix product operators, MPO) to explicitly construct arbitrary many - body Hamiltonians. For two - body interactions, this method is always effective, and the required tensor dimension grows linearly with the number of particles. For systems with specific symmetries or constraints, an optimal representation with the minimum tensor dimension can be found. In addition, the paper also discusses how to analytically and numerically approximate operators in low - dimensional tensor operators and shows its applications in time - evolution and ground - state approximation, especially in non - uniformly coupled long - range interaction systems. These operator representations are also extended to other geometric structures, such as tree - like and two - dimensional lattices, showing how to obtain and use efficient tensor network representations while respecting the given geometric structure.
Specifically, the paper addresses the following aspects:
1. **Effectively representing operators of long - range interaction systems**: A method of using matrix product operators (MPO) to represent arbitrary many - body Hamiltonians is proposed, which is especially suitable for long - range interaction systems.
2. **Optimizing tensor dimension**: For systems with specific symmetries or constraints, an optimal representation with the minimum tensor dimension is found.
3. **Analytical and numerical approximation**: How to analytically and numerically approximate operators in low - dimensional tensor operators is discussed.
4. **Application examples**: The application of this method in time - evolution and ground - state approximation is shown, especially in non - uniformly coupled long - range interaction systems.
5. **Extension to geometric structures**: These operator representations are extended to other geometric structures, such as tree - like and two - dimensional lattices, showing how to obtain and use efficient tensor network representations while respecting the given geometric structure.
Through these methods, the paper provides a systematic way to construct and optimize tensor network representations, thereby improving computational efficiency when dealing with complex quantum systems.