Unitary-Coupled Restricted Boltzmann Machine Ansatz for Quantum Simulations

Chang-yu Hsieh,Qiming Sun,Shengyu Zhang,Chee Kong Lee

DOI: https://doi.org/10.48550/arXiv.1912.02988

2019-12-06

Abstract:Neural-Network Quantum State (NQS) has attracted significant interests as a powerful wave-function ansatz to model quantum phenomena. In particular, a variant of NQS based on the restricted Boltzmann machine (RBM) has been adapted to model the ground state of spin lattices and the electronic structures of small molecules in quantum devices. Despite these progresses, significant challenges remain with the RBM-NQS based quantum simulations. In this work, we present a state-preparation protocol to generate a specific set of complex-valued RBM-NQS, that we name the unitary-coupled RBM-NQS, in quantum circuits. This is a crucial advancement as all prior works deal exclusively with real-valued RBM-NQS for quantum algorithms. With this novel scheme, we achieve (1) modeling complex-valued wave functions, (2) using as few as one ancilla qubit to simulate $M$ hidden spins in an RBM architecture, and (3) avoiding post-selections to improve scalability.

Quantum Physics

What problem does this paper attempt to address?

This paper aims to address several key challenges encountered in quantum simulation when using neural - network quantum states (NQS) as variational wave functions. Specifically, the paper attempts to solve the following problems:

1. **Reduce qubit resource consumption**: Existing NQS methods based on restricted Boltzmann machines (RBM) need to explicitly model all visible and hidden spins, which leads to the need for a large number of ancilla qubits. These qubits are very precious resources on current noisy intermediate - scale quantum (NISQ) devices. The paper proposes a new state - preparation protocol that can simulate multiple hidden spins with only one ancilla qubit, thereby greatly reducing qubit consumption.

2. **Avoid post - selection**: In order to achieve non - unitary transformations, existing NQS methods usually rely on probabilistic post - selection techniques. This not only reduces the efficiency of the algorithm but also limits its application in large - scale systems. The paper proposes a new scheme that completely avoids post - selection. By decomposing a single NQS into a set of modified NQS, it is possible to extract the observable estimates of the original NQS from the set of "post - selection failures" in the Monte Carlo framework.

3. **Support complex wave functions**: Many quantum systems, such as fermionic systems in the presence of a magnetic field and quantum - dynamic simulations, require complex wave functions for accurate description. However, existing NQS methods mainly deal with real - valued wave functions, limiting their scope of application. The paper proposes a method for generating complex NQS and shows how to achieve this goal in a quantum circuit.

4. **Improve algorithm efficiency**: For large - scale simulations, the number of parameters \(N_{\text{var}}\) in NQS can be very large. The standard imaginary - time - dependent variational principle (ITDVP) method needs to prepare \(O(N_{\text{var}}^2)\) quantum states to calculate matrix elements, which is not feasible in practice. The paper proposes an improved ITDVP algorithm that only needs to perform measurements on a single quantum circuit, greatly simplifying the calculation process.

Through the above innovations, the paper expands the application range of NQS in hybrid quantum - classical algorithms and lowers the threshold for its experimental implementation, providing new possibilities for future quantum computing applications in complex quantum - system simulations.