Universal Hamming Weight Preserving Variational Quantum Ansatz

Understanding the mathematical properties of variational quantum ans\"atze is crucial for determining quantum advantage in Variational Quantum Eigensolvers (VQEs). A deeper understanding of ans\"atze not only enriches theoretical discussions but also facilitates the design of more efficient and robust frameworks for near-term applications. In this work, we address the challenge of balancing expressivity and trainability by utilizing a Hamming Weight Preserving (HWP) ansatz that confines quantum state evolution to a symmetry-preserving subspace. We rigorously establish the necessary and sufficient conditions for subspace universality of HWP ans\"atze, along with a comprehensive analysis of the trainability. These theoretical advances are validated via the accurate approximation of arbitrary unitary matrices in the HWP subspace. Furthermore, the practical utility of the HWP ansatz is substantiated for solving ground-state properties of Fermionic systems, achieving energy errors below \(1\times 10^{-10}\)Ha. This work highlights the critical role of symmetry-preserving ans\"atze in VQE research, offering insights that extend beyond supremacy debates and paving the way for more reliable and efficient quantum algorithms in the near term.