A Euclidean Monte-Carlo-informed route to ground-state preparation for quantum simulation of scalar field theory
Zohreh Davoudi · 2025
Read A Euclidean Monte-Carlo-informed route to ground-state preparation for quantum simulation of scalar field theory by Zohreh Davoudi (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2510-24875.txt.
Introduction
Navya Gupta, Christopher David White, and Zohreh Davoudi bridge two views of a field theory to set up states for a quantum run. Real time acts are what quantum tools promise, but hard start states block the way. Classic Monte Carlo runs in fake time give ground state facts that could guide that setup.
The team works on a one space plus one time field with self push. They build trial state forms that hit like ground energies yet differ in links and local shape. By tuning set shape stats to match Monte Carlo data, they get forms that turn into quantum circuits with ease. Cost grows in a tame, power like way with size.
Catalog record
| CREATOR | Zohreh Davoudi |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 22 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2510-24875
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
A Euclidean Monte-Carlo-informed route to ground-state preparation for quantum simulation of scalar field theory Zohreh Davoudi Abstract Quantum simulators hold great promise for studying real-time (Minkowski) dynamics of quantum field theories. Nonetheless, preparing non-trivial initial states remains a major obstacle. Euclidean-time Monte-Carlo methods yield ground-state spectra and static correlation functions that can, in principle, guide state preparation. In this work, we exploit this classical information to bridge Euclidean and Minkowski descriptions for a (1+1)-dimensional interacting scalar field theory. We propose variational ansatz families (inspired by the stellar hierarchy for bosonic systems introduced in Ref. [ 1 ] ) which achieve comparable ground-state energies, yet exhibit distinct correlations and local non-Gaussianity. By optimizing selected wavefunction moments with Monte-Carlo data, we obtain ansatzes that can be efficiently translated into quantum circuits. Our algorithmic cost analysis shows these circuits’ gate complexity scales polynomially in system size. Our work paves the way for systematically leveraging classically-computed information to prepare initial states in quantum field theories of interest in nature. [a,b,c]Navya Gupta 1 Introduction Real-time (Minkowski) dynamics of quantum field theories (QFTs) are notoriously difficult to address with standard Euclidean lattice methods due to a sign problem [ 2 , 3 , 4 ] . Quantum simulators offer the exciting possibility of studying such dynamics [ 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 ] . However, a major bottleneck is efficiently preparing non-trivial initial states, such as ground or low-lying excited states [ 15 , 16 , 17 , 18 , 19 ] . At the same time, path-integral Monte-Carlo (PIMC) techniques in Euclidean spacetime remain highly effective at extracting static observables in QFTs, including correlation functions and spectra [ 20 , 21 , 22 , 23 , 24 ] (when no sign or signal-to-noise problem hinders the computation). This raises the prospect of exploiting information about ground-state correlation functions obtained via Euclidean methods to guide or improve state preparation on quantum computers which operate in Minkowski time. In this work, we demonstrate a strategy to bridge these two pictures in the context of an interacting scalar field theory in (1+1) dimensions (see Fig. 1 ). Specifically, we use PIMC data for ground-state correlation functions to inform the optimization of a variational family of wavefunction ansatzes. This variational ansatz is inspired by the stellar hierarchy for bosonic quantum systems [ 1 , 25 ] . The knowledge of ground-state correlation functions (i.e., moments) allows us to jointly optimize the energy and moments of this ansatz. Furthermore, we show that the ansatz admits an efficient mapping to a quantum circuit. This allows us to “compile” the optimized ansatz into a circuit whose gate complexity is polynomial in the lattice size and can be implemented on a qubit-based quantum computer. Thus, this method relies solely upon classical computing to determine the quantum circuit for implementing the ground state. Once this circuit is realized on quantum hardware, it can be used as an input for the quantum computation of dynamical correlation functions. In summary, the two main pillars of our work are: 1) A bosonic ansatz that is classically tractable (i.e, can be optimized efficiently using classical resources), circuit translatable (i.e., can be efficiently translated into a quantum circuit using classical resources), and circuit efficient (i.e, the resulting quantum circuit’s complexity scales polynomially with system size); and 2) a Euclidean-Monte-Carlo-informed moment-optimization procedure for optimizing this ansatz. The efficiency of our ansatz stems from the fact that it is directly expressed in terms of the degrees of freedom of the target field theory, making it straightforward to encode features of the theory (e.g., correlations, non-Gaussianity) in the ansatz. Given that any ansatz will only express limited features of a non-trivial interacting ground state, we use Euclidean-Monte-Carlo-informed moment optimization to tune the ansatz to reproduce the most relevant features of the ground state. Such features depend on the subsequent quantum-simulation goals, e.g., studies of certain excitations and their dynamics. Figure 1 : Schematic overview of classically informed ground-state preparation. Using ground-state correlation functions in an interacting scalar field theory, sourced from Euclidean path-integral Monte Carlo, a joint optimization of the energy and moments of the ansatz wavefunction is performed. Based on a mapping between the field and simulator degrees of freedom, the optimized ansatz is translated into a quantum circuit using a classical algorithm. This classically determined circuit can thereafter be implemented on quantum hardware, and be used as the starting point for performing other tasks, such as simulating real-time dynamics and estimating dynamical correlation functions. The goal of the work presented is to demonstrate a classical determination of the quantum circuit for preparing the ground state. 2 The Model: (1+1)D \phi^{4} theory The Hamiltonian for the continuum (1+1)D lattice \phi^{4} theory is given by \hat{H}_{\mathrm{cont}}\;=\;\int\!dx\,\Bigl[\tfrac{1}{2}\,\hat{\pi}(x)^{2}\;+\;\tfrac{1}{2}\,\bigl(\partial_{x}\hat{\phi}(x)\bigr)^{2}\;+\;\tfrac{1}{2}\,m_{0}^{2}\,\hat{\phi}(x)^{2}\;+\;\tfrac{\lambda_{0}}{4}\,\hat{\phi}(x)^{4}\Bigr], (1) where m_{0}^{2} and \lambda_{0} are the bare mass and coupling, respectively, and the fields \hat{\phi} and \hat{\pi} satisfy the canonical bosonic commutation relation [\hat{\phi}(x),\hat{\pi}(x^{\prime})]=i\,\delta(x-x^{\prime}) . We discretize space to a periodic lattice with spacing a and length L=Na . Defining the dimensionless fields \hat{\phi}_{j}\coloneq\hat{\phi}(ja) and \hat{\pi}_{j}\coloneq a\,\hat{\pi}(ja) , the dimensionless lattice Hamiltonian reads \displaystyle\hat{H}\;\coloneq\;a\,\hat{H}_{a,N}\;=\;\sum_{j=0}^{N-1}\Bigl[\tfrac{1}{2}\,\hat{\pi}_{j}^{2}\;+\;\tfrac{1}{2}\,\bigl(\hat{\phi}_{j+1}-\hat{\phi}_{j}\bigr)^{2}\;+\;\tfrac{1}{2}\,m^{2}\,\hat{\phi}_{j}^{2}\;+\;\tfrac{\lambda}{4}\,\hat{\phi}_{j}^{4}\Bigr], (2) where m^{2}\coloneq m_{0}^{2}a^{2} and \lambda\coloneq\lambda_{0}a^{2} . This Hamiltonian possesses ( \mathbb{Z}_{2} ) parity [i.e., (\hat{\phi},\hat{\pi})\mapsto(-\hat{\phi},-\hat{\pi}) ], time reversal, lattice inversion (i.e., spatial parity), and ( \mathbb{Z}_{N} ) lattice translation symmetries. For m^{2}<0 , the parity symmetry is spontaneously broken for a critical value of \lambda in the thermodynamic limit [ 26 ] . The continuum limit is given by taking N\rightarrow\infty followed by (m^{2},\lambda)\rightarrow(0,0) . For this limit to converge, the couplings (m^{2},\lambda) must be tuned suitably with a [ 15 ] . To keep our PIMC simple, we work with a modest lattice size N=10 . We choose (m^{2},\lambda) to ensure the lightest scalar mode fits well within the system volume ( N\Delta E>1 ) and remains below the UV cutoff ( \Delta E<\pi ), where \Delta E denotes the spectral gap. Instead of determining the exact tuning of the couplings (m^{2},\lambda) with the lattice spacing a , we simply take a series of four values of (m^{2},\lambda) approaching (0,0) along a straight line in the symmetric phase. PIMC is then used to estimate the two-point function \langle\hat{\phi}_{0}\,\hat{\phi}_{j}\rangle , which characterizes non-local correlations in the ground state. Additionally, the moment ratio R_{2n}\;\coloneq\;\frac{\bigl\langle\hat{\phi}_{j}^{2n}\bigr\rangle}{(2n-1)!!\,\bigl\langle\hat{\phi}_{j}^{2}\bigr\rangle^{n}}, (3) a simple measure of the local non-Gaussianity of the ground state, is computed from local moments. 3 The (R,Q) ansatz. The ladder operators for the N bosonic modes are given by \hat{a}_{j}\;\coloneq\;\frac{\hat{\phi}_{j}\,+\,i\,\hat{\pi}_{j}}{\sqrt{2}} and \hat{a}_{j}^{\dagger}\;\coloneq\;\frac{\hat{\phi}_{j}\,-\,i\,\hat{\pi}_{j}}{\sqrt{2}} ; they satisfy the commutation relations [\hat{a}_{j},\hat{a}_{k}^{\dagger}]\;=\;\delta_{j,k} and [\hat{a}_{j},\hat{a}_{k}]\;=\;0 . An N -mode bosonic state |\psi\rangle is said to have a finite stellar rank R if it can be written as [ 1 , 25 ] |\psi\rangle\;=\;\hat{U}_{G}\,|C\rangle. (4) Here, \hat{U}_{G} is a Gaussian unitary (i.e., is generated by a quadratic Hamiltonian in \hat{a}_{j} and \hat{a}_{j}^{\dagger} ). Furthermore, |C\rangle\coloneq C(\hat{a}_{0}^{\dagger},\cdots,\hat{a}_{N-1}^{\dagger})\,|\mathbf{0}\rangle is referred to as the core state : it is generated from the Fock vacuum |\mathbf{0}\rangle\coloneq|0,\cdots,N-1\rangle by C(\hat{a}_{0}^{\dagger},\cdots,\hat{a}_{N-1}^{\dagger}) —a polynomial in the boson creation operators with degree R . Bosonic states which do not admit the decomposition given in Eq. ( 4 )—such as the \phi^{4} theory’s ground state—are said to have an infinite rank. Finite-rank states form a