arXiv:2610.09532
Vibrational Sample-Based Quantum Diagonalization with Unitary-Cluster-Jastrow Ansätze on IBM QPUs
Kunal Kumar, Erik Lötstedt, Yuri Kobayashi, Seiji Yunoki
Abstract
We extend sample-based quantum diagonalization (SQD) to anharmonic vibrational spectra and use it to benchmark how vibrational ansätze scale on present-day hardware under a direct one-hot encoding (up to 72 qubits). Five particle-conserving circuits are warm-started through vibrational coupled-cluster (VCC) t1, t2 amplitudes: UVCCSD, the compact heuristic circuit (CHC), and three unitary-cluster-Jastrow ansätze introduced here, VLUCJ, Vg-uCJ, and VIm-uCJ. A self-consistent configuration-recovery loop reconstructs correlated states from hardware samples in the low-retention regime where naïve post-selection breaks down, enabling the calculation of ground-state energies for H$_2$O, CH$_2$O, and CH$_2$ClF (3-9 modes). We compare harmonic-oscillator and vibrational self-consistent-field (VSCF, "modal") reference bases. In the small-modal-basis regime relevant for near-term devices, the modal basis is typically more accurate than the harmonic-oscillator representation; empirically, it also yields substantially better-converged hardware sampling for deep circuits at these qubit counts. Across ansätze, we find a clear retention/accuracy tradeoff when scaling: the new VLUCJ circuit is the most hardware-robust, retaining the largest fraction of physically valid one-hot outcomes, whereas the more expressive uCJ variants and UVCCSD can recover more correlation when retention is sufficient. Finally, we extend the recovery protocol to excited states by targeting a chosen per-mode VSCF occupation and warm-starting it with a state-specific VCC calculation, and we demonstrate the calculation of one representative fundamental vibrational frequency per molecule on IBM QPUs.
AI Summary
The paper asks how sample-based quantum diagonalization (SQD) can be adapted to anharmonic molecular vibrations, and which sampling circuits remain useful under hardware noise. It uses a direct one-hot modal encoding, classical vibrational coupled-cluster warm starts, and self-consistent configuration recovery: invalid mode blocks are repaired using modal occupations estimated from successive subspace diagonalizations. Five ansätze—UVCCSD, CHC, and three vibrational cluster-Jastrow variants—are compared in harmonic-oscillator and VSCF-modal bases. Hardware experiments cover H$_2$O, CH$_2$O, and CH$_2$ClF, with 3–9 modes, 2–8 modals per mode, and 6–72 qubits on two IBM processors. VLUCJ has the highest average valid one-hot retention: 25.6% and 54.7% on the respective backends. Retention and recovered-energy accuracy do not rank circuits identically. On the lower-noise backend, the other four ansätze have median ground-energy deviations of $0.27$–$1.40\,\mathrm{cm}^{-1}$ from the classical VCC hierarchy, versus $2.0\,\mathrm{cm}^{-1}$ for VLUCJ. State-specific warm starts and reference-overlap root selection also recover one fundamental frequency per molecule; UVCCSD’s median frequency deviation is $0.21\,\mathrm{cm}^{-1}$. Validation uses VFCI where available, VCC/VHCI references, and uniform valid-configuration sampling controls. Uniform sampling fails to reproduce the larger-system results, but does not exclude structured hardware bias. Several large circuits yield no valid raw shots, limiting interpretation of their recovered energies. No quantum advantage is claimed; repeated-run uncertainty analysis, broader spectra, transition intensities, and scaling beyond 72 qubits remain open.
Error found: The specification that every VIm-uCJ and Vg-uCJ production circuit uses P=8 distinct selected pairs is impossible for H2O at two modals per mode, where only three such pairs exist. The stated HO-to-modal transformation uses adjoints in the wrong direction for the defined M-by-K coefficient matrices, so its Hamiltonian-transformation formula is dimensionally invalid when M differs from K.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.