Max-Cut graph-driven quantum circuit design for planar spin glasses
Seyed Ehsan Ghasempouri, Gerhard W. Dueck, Stijn De Baerdemacker · 2025
Read Max-Cut graph-driven quantum circuit design for planar spin glasses by Seyed Ehsan Ghasempouri, Gerhard W. Dueck, Stijn De Baerdemacker (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-2504-12096.txt.
Introduction
Seyed Ehsan Ghasempouri, Gerhard Dueck, and Stijn De Baerdemacker design quantum circuits in 2025 for frustrated magnet systems. In such systems, competing pulls between spins create rugged energy landscapes where the lowest state hides. Finding that ground state matters for many hard problems recast in this form.
Working with flat lattices built from triangular units, the team initializes states for up to twenty sites without stalling training. A graph cut splits the lattice into two maximally apart groups, which guides an efficient circuit layout. The work runs within the standard hybrid quantum-classical solver of the day. Costs stay polynomial, the authors show.
Catalog record
| CREATOR | Seyed Ehsan Ghasempouri, Gerhard W. Dueck, Stijn De Baerdemacker |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 25 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-12096
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Max-Cut graph-driven quantum circuit design for planar spin glasses Preprint: APS/123-QED Seyed Ehsan Ghasempouri, Gerhard W. Dueck, Stijn De Baerdemacker Abstract Finding the ground state of spin glasses is a challenging problem with broad implications. Many hard optimization problems, including NP-complete problems, can be mapped, for instance, to the Ising spin glass model. We present a graph-based approach that allows for accurate state initialization of a frustrated triangular spin-lattice with up to 20 sites that stays away from barren plateaus. To optimize circuit efficiency and trainability, we employ a clustering strategy that organizes qubits into distinct groups based on the maximum cut technique, which divides the lattice into two subsets maximally disconnected. We provide evidence that this Max-Cut-based lattice division offers a robust framework for optimizing circuit design and effectively modeling frustrated systems at polynomial cost. All simulations are performed within the variational quantum eigensolver (VQE) formalism, the current paradigm for noisy intermediate-scale quantum (NISQ), but can be extended beyond. Our results underscore the potential of hybrid quantum-classical methods in addressing complex optimization problems. I Introduction Determining the ground state of an Ising spin glass edwards1975theory , a problem within the class of combinatorial optimization, becomes exponentially challenging for classical computers as the system size increases, similar to other many-body problems. Quantum computers, first proposed in the early 1980s Benioff ; Manin ; Feynman , are anticipated to surpass these limitations by leveraging quantum mechanical phenomena, including superposition, interference, and entanglement. This advancement in quantum computing offers significant potential for solving specific computational challenges. Two of the most prominent algorithms in this field are the quantum phase estimation (QPE) Kitaev ; Kitaev_Alexei ; Cleve ; Nielsen ; Abrams and the variational quantum eigensolver (VQE) Peruzzo ; McClean . The QPE algorithm typically requires a high circuit depth to accurately determine the ground state energy, which necessitates further development in error-corrected hardware. Consequently, the VQE, a quantum-classical hybrid approach with shallow circuit depth, is a viable alternative for noisy intermediate-scale quantum (NISQ) devices Preskill . However, this advantage comes at the cost of potentially encountering an NP-hard optimization problem during the training process bittel2021training . The objective of the VQE is to find a parameterized ansatz that can determine the energy of the target state or provide a reasonable quantum state estimation for the QPE. The ansatz is prepared by designing a quantum circuit with a variational form, consisting of layers that specify rotation patterns and CNOT gates. One of the most challenging aspects of VQE is the design of an efficient variational quantum circuit with shallow depths, which is crucial in noisy environments and demands a minimal number of parameters to avoid issues related to barren plateaus and entrapment in local minima. The Ising problem has a tight connection with optimization problems. The Quantum Approximate Optimization Algorithm (QAOA), introduced by Farhi, Goldstone, and Gutmann in 2014 farhi2014quantum , is an optimization algorithm based on the VQE method designed to address combinatorial problems. The QAOA has been widely studied to solve quadratic unconstrained binary optimization (QUBO) problems BLEKOS20241 ; moussa2022unsupervised such as Maximum Cut (Max-Cut) farhi2014quantum ; basso2021quantum , Maximum Independent Set (MIS) choi2019tutorial , Binary Paint Shop Problem (BPSP) streif2021beating , Binary Linear Least Squares (BLLS) borle2021quantum , etc. The QAOA method has many scientific and industrial applications such as portfolio optimization hodson2019portfolio , tail assignment vikstaal2020applying , protein folding mustafa2022variational , etc. In several situations, such as QUBO, the QAOA can be mapped to the problem of finding the ground state of an Ising system. Exploring the ground-state properties of Ising spin glasses stands as a key difficulty in both solid-state physics and statistical mechanics and has been extensively researched for many decades binder1986spin ; mezard1987spin ; mezard2009information ; stein2013spin . The term spin glass denotes a state of disordered magnetism where the spins of magnetic atoms or molecules interact in such a way that it gives rise to frustrated magnetic phenomena mydosh1993spin . The behavior of frustrated magnetic metals is observed in alloys such as CuMn and AuFe, where manganese (Mn) or iron (Fe) is introduced as impurities into copper (Cu) or gold (Au) cannella1972magnetic ; mulder1981susceptibility . The Ising model is widely used due to its simplicity and effectiveness in capturing the essential features of real spin glasses, as it accounts for the interaction energy between impurities de1995exact ; barahona1982computational . The presence of unsatisfied edges in the lattice can lead to situations where multiple geometric arrangements of atoms remain stable. For a spin glass with n spins, there are 2^{n} possible configurations, making the task of finding the ground state a combinatorial problem. It has been shown that for planar graphs with nearest-neighbor interactions, both with and without periodic boundary conditions, and in the absence of an external magnetic field, the ground state can be found in polynomial time edmonds1965minimum ; bieche1980ground ; barahona1982morphology ; d1984random ; hadlock1975finding ; orlova1972finding and is comparable to solving the Max-Cut problem. However, Barahona proved that computing the ground state energy for two-dimensional (2D) models with a magnetic field is NP-hard, even without periodic boundary conditions, and no polynomial algorithm exists for this problem barahona1982computational . Over the past few decades, various methods, such as Monte Carlo simulations de1996exact and evolutionary algorithms gropengiesser1995ground ; gropengiesser1995superlinear , including genetic algorithms sutton1994genetic , have been developed to determine the ground state energy of Ising spin glasses. However, these approaches generally provide only approximations rather than the exact ground state, with a significant limitation that there is no guarantee of finding the true ground state barahona1994ground ; de1995exact . Additionally, another major drawback of these heuristic methods is their performance in the presence of degeneracies. Two states with almost identical energy levels can still be qualitatively different. Therefore, even though these algorithms can give energies close to the ground state, they may not be close to the true ground state. As a result, these methods are inadequate for analyzing real ground states, such as those found in protein folding applications, as mentioned in reference liang1992application . An alternative approach to finding the true ground state involves adiabatic methods, such as simulated annealing (SA) kirkpatrick1983optimization , quantum annealing (QA) johnson2011quantum , and the quantum adiabatic algorithm (QAA) farhi2000quantum . The SA method is an optimization technique that seeks the ground state by simulating a system gradually cooling into a low-energy state. The downside of this method is that the cooling process must be prolonged; otherwise, the system may get trapped in local minima rather than reaching the global minimum. The QA relies on quantum fluctuations and aims to find optimal solutions by gradually reducing an applied transverse magnetic field to the spin system at the lowest possible temperature. Experiments brooke1999quantum and simulations kadowaki1998quantum ; santoro2002theory have demonstrated that QA can achieve equilibrium in a spin glass more rapidly than SA, which operates through thermal annealing. However, removing the magnetic field reduces quantum fluctuations and forces the system to pass through the critical point. A complete understanding of the transition between the disordered phase, which occurs at large fields, and the spin glass phase, which emerges at smaller fields, is still lacking and remains an active area of research. A recent study bernaschi2024quantum utilizing extreme-scale