Abstract:The real- and imaginary-time evolution of quantum states are powerful tools in physics, chemistry, and beyond, to investigate quantum dynamics, prepare ground states or calculate thermodynamic observables. On near-term devices, variational quantum time evolution is a promising candidate for these tasks, as the required circuit model can be tailored to trade off available device capabilities and approximation accuracy. However, even if the circuits can be reliably executed, variational quantum time evolution algorithms quickly become infeasible for relevant system sizes due to the calculation of the Quantum Geometric Tensor (QGT). In this work, we propose a solution to this scaling problem by leveraging a dual formulation that circumvents the explicit evaluation of the QGT. We demonstrate our algorithm for the time evolution of the Heisenberg Hamiltonian and show that it accurately reproduces the system dynamics at a fraction of the cost of standard variational quantum time evolution algorithms. As an application of quantum imaginary-time evolution, we calculate a thermodynamic observable, the energy per site, of the Heisenberg model.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the computational bottleneck encountered by the Variational Quantum Time Evolution (VarQTE) algorithm in handling relevant system scales in quantum computing, especially the problem of excessive resource requirements due to the need to calculate the Quantum Geometric Tensor (QGT). Specifically, as the system scale increases, the number of variational parameters increases, and the number of circuits required to calculate the QGT grows in proportion to \(O(d^2)\), which makes the algorithm impractical in practical applications. To solve this problem, the paper proposes a new method - the Dual Variational Quantum Time Evolution (DualQTE) algorithm. By introducing a dual - optimization problem to avoid directly calculating the QGT, it significantly reduces the required resources and improves the efficiency and feasibility of the algorithm.
### Core improvement points of the algorithm
1. **Dual - optimization problem**: The paper proposes a dual - optimization problem. By minimizing the loss function \(L(\delta\theta)\), where the loss function is defined as:
\[
L(\delta\theta)=\frac{1 - F(\theta,\theta+\delta\theta)}{2}-\delta\tau\cdot\delta\theta^{T}b(\theta)
\]
Here, \(F(\theta,\theta')\) is the fidelity between the variational states corresponding to the two parameter sets, \(b(\theta)\) is the evolution gradient, and \(\delta\tau\) is the time perturbation. This method avoids directly calculating the QGT and instead achieves parameter updates by evaluating the fidelity.
2. **Reduction of resource requirements**: Compared with the traditional VarQTE, DualQTE only requires \(O(d)\) resources to calculate the evolution gradient \(b\) and fidelity \(F\) at each time step, instead of \(O(d^2)\) resources to calculate the QGT. In addition, the number of optimization steps \(K\) is usually much smaller than \(d\), further reducing the resource requirements.
3. **Numerical stability**: The paper uses the gradient - descent method to solve the dual - optimization problem and utilizes the Parameter - Shift Rule or the Linear Combination of Unitaries (LCU) to calculate the fidelity and evolution gradient, improving the numerical stability.
### Experimental verification
The paper verifies the effectiveness of DualQTE by simulating the imaginary - time evolution of the Heisenberg Model. The experimental results show that even when there are only 100 measurements in each circuit evaluation, DualQTE can qualitatively track the imaginary - time evolution, and in some cases is even better than VarQTE with 1024 measurements. When the number of measurements is increased to 1024, DualQTE can more accurately approach the exact solution, showing higher precision.
### Conclusion
The DualQTE algorithm proposed in the paper effectively solves the computational bottleneck problem of the traditional VarQTE algorithm in handling large - scale systems by introducing a dual - optimization problem, significantly reduces the resource requirements, and improves the feasibility and efficiency of the algorithm. This improvement is of great significance for achieving efficient quantum time evolution on near - term quantum devices.