Efficient quantum state preparation through seniority driven operator selection

Dipanjali Halder 1 , Dibyendu Mondal 1 , Rahul Maitra 1,2,† · 2025

Read Efficient quantum state preparation through seniority driven operator selection by Dipanjali Halder 1 , Dibyendu Mondal 1 , Rahul Maitra 1,2,† (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-2504-19760.txt.

Introduction

Dipanjali Halder, Dibyendu Mondal, and Rahul Maitra prepare quantum states for strongly correlated molecules with lean circuits, in work from 2025. Quantum hardware needs compact recipes for electron states, but accuracy and gate cost fight each other. Many top methods also demand heavy measurement before the circuit even runs.

Their framework builds the state from ranked excitation operators chosen by seniority. Low-rank, seniority-zero moves capture strong correlation at shallow depth. A hybrid pruning step, part intuition and part testing, trims weak excitations. Pre-circuit measurement overhead drops as a result. The balance struck suits near-term quantum devices. Chemistry with stubborn electron correlation is the target use.

Catalog record

CREATORDipanjali Halder 1 , Dibyendu Mondal 1 , Rahul Maitra 1,2,†
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE51 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-19760

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Efficient quantum state preparation through seniority driven operator selection Dipanjali Halder 1 , Dibyendu Mondal 1 , Rahul Maitra 1,2,† Abstract Quantum algorithms require accurate representations of electronic states on a quantum device, yet the approximation of electronic wave functions for strongly correlated systems remains a profound theoretical challenge, with existing methods struggling to balance the competing demands of chemical accuracy and gate efficiency. Moreover, a critical limitation of the most of the state-of-the-art methods developed to date lies in their substantial reliance on extensive pre-circuit measurements, which introduce significant overheads and contribute to inefficiencies in practical implementation. To address these interconnected challenges and establish a harmonious synergy between them, we propose an algorithmic framework that focuses on efficiently capturing the molecular strong correlation through an ordered set of computationally less demanding rank-one and seniority-zero excitations, yielding a parameterized ansatz with shallow gate depth. Furthermore, to achieve minimal pre-circuit measurement overhead, we implement a selective pruning of excitations through a hybrid approach that combines intuition-based selection with shallow-depth, rank-one excitations driven uni-parameter circuit optimization strategy. With the incorporation of qubit-based excitations via particle-preserving exchange circuits, we demonstrate a further reduction in quantum complexities, enhancing the overall resource efficiency of the approach. With a range of challenging applications on strongly correlated systems, we demonstrate that our dynamic ansatz not only significantly enhances computational efficiency but also delivers exceptional accuracy, robustness, and resilience to the noisy environments inherent in near-term quantum hardware. \nolinenumbers I Introduction Solving the electronic Schrödinger equation lies at the heart of theoretical chemistry, offering critical insights into molecular structure, behavior, and interactions. Traditional classical methods, however, are severely limited by the curse of dimensionality, as the computational cost grows exponentially with the system size. Quantum computing offers a compelling alternative to address these exponential bottlenecks [ 1 , 2 ] , though its practical realization remains constrained by short coherence times, gate imperfections, and limited circuit depths. To navigate these limitations, quantum-classical hybrid algorithms, notably the Variational Quantum Eigensolver (VQE) [ 3 ] have emerged as a promising pathway toward achieving quantum advantage in quantum chemistry. The feasibility of such hybrid algorithms on the near term quantum devices, however, relies on the ability of approximate wavefunction ansatzes to produce accurate energies and properties while minimizing the quantum complexities. Generally, the wavefunction ansatzes are typically constructed using either hardware-efficient or chemically motivated fermionic operators, each with distinct trade-offs [ 4 , 5 , 6 ] . Hardware efficient ansatz, specifically designed to optimize the use of available quantum resources, offers a more pragmatic approach but fails to preserve physical symmetries and faces scalability limitations due to optimization challenges. Whereas the chemically motivated ansatzes, such as those based on traditional unitary coupled cluster (UCC) method or other problem-specific methods, inherently preserve physical symmetries, enable more straightforward optimization, and allow systematic improvement through higher-order excitations. However, they tend to be more resource-intensive, often demanding significantly deeper and complex circuits. Hence, the challenge lies in balancing the ansatz’s capability and flexibility to capture complex many-electron correlations with the inherent limitations of near-term quantum hardware. In addressing this challenge, significant progress has been made toward developing efficient and physically meaningful ansatzes. Among the earliest and most notable approaches is the unitary coupled cluster singles and doubles (UCCSD) method [ 3 , 7 ] , which had set a foundational framework for further advancements in the field. However, due to the non-commuting nature of the operator components within UCCSD, the Trotterized implementation of this method is inherently non-unique, leading to substantial variations in accuracy across different operator orderings [ 8 ] . Moreover, as the complexity of molecular correlations escalates, Trotterized UCCSD may fail to achieve chemical accuracy, often incorporating numerous unnecessary excitations that do not contribute meaningfully to the correlation, yet significantly inflate the circuit depth, thereby exacerbating the challenges of efficient quantum simulation. Recent investigations have extended UCCSD by incorporating operators beyond conventional particle-hole excitations [ 9 , 10 , 11 ] . Notable among these advancements is k-UpCCGSD [ 9 ] , which achieves linear scaling by employing paired generalized rank-two excitations combined with generalized single excitations. As quantum systems grow in size and complexity, fixed ansatzes like k-UpCCGSD often fail to maintain the delicate balance between accuracy and resource efficiency. Consequently, it becomes imperative to develop dynamic approaches that can effectively tailor the ansatz as per system-specific criteria, ensuring both accuracy and computational feasibility. A prominent example of such a state-of-the-art approach is ADAPT-VQE [ 12 ] , which efficiently extracts correlation energy using a dynamically identified, problem specific minimal subset of operators. However, its reliance on gradient-based operator selection introduces substantial measurement overhead, which scales unfavorably with the system size. Unlike the ADAPT-VQE approach, an alternative and notable approach in this context is the DISCretely Optimized Variational Quantum Eigensolver (DISCO VQE) [ 13 ] which introduces a distinct strategy by incorporating products of spin symmetry-preserving fermionic operators, while simultaneously optimizing both the sequence of operators and their amplitudes. Addressing the high measurement overhead associated with adaptive algorithms, the present authors have introduced COMPASS-VQE [ 14 ] , a protocol that identifies dominant operators using parallel one- and two-parameter circuits, thus reducing resource demands. The same set of authors have subsequently introduced two distinct ansatz compactification strategies: one grounded in first-principles-based many-body perturbation theory (COMPACT-VQE) [ 15 ] and the other driven by unsupervised machine learning techniques (RBM-VQE) [ 16 ] . For further insights and related advancements in this area, the reader is referred to the literatures cited herein [ 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 ] . Despite these advancements, the accurate representation of electronic wavefunctions at low computational cost remains a major challenge in the field, highlighting the importance of continued development of efficient and scalable methods. While highly accurate approaches often rely on the inclusion of intricate and computationally prohibitive higher-rank excitations, it is well established that the number of two-qubit CNOT gates required to implement an n -tuple fermionic or qubit excitation scales exponentially with n [ 29 ] . Given that two-qubit gates, such as CNOTs, are significantly more error-prone (often by an order of magnitude) than the single-qubit gates, they inevitably constitute the primary source of gate errors in practical quantum hardware. Consequently, reducing the number of CNOT gates becomes imperative for realizing accurate and experimentally feasible quantum simulations. A conceivable route to address this challenge lies in the strategic parametrization of the electronic wavefunction, where the use of computationally efficient low-rank excitations may suffice to capture essential correlation effects, thereby circumventing the need for explicitly incorporating higher-rank excitations. In consonance with this idea, we propose an algorithm that relies predominantly on the computationally less demanding rank-one excitations, which serve as pivotal elements capable of spanning higher seniority sectors of the Hilbert space when supplemented by a sparse subset of rank-two, seniority-preserving paired excitations, thereby achieving an optimal balance between computational tractability and accurate description of strong electron correlations. In addition to the use of structurally simple and computationally efficient operator pool, our method dynamically identifies the most significant and chemically relevant operators from the pool,

Read the complete text