Machine-Learned Compact Subspace Generation for Quantum Selected Configuration Interaction within Density Matrix Embedding Framework
Abstract
Sample-based quantum diagonalization (SQD), an extension of quantum selected configuration interaction (QSCI), is a promising hybrid quantum-classical paradigm for molecular ground-state energies. By replacing variational optimization with quantum sampling, QSCI avoids barren plateaus and directly reconstructs correlated wavefunctions. However, existing configuration recovery mainly enforces symmetry constraints without targeting the most physically relevant configurations, often yielding oversized subspaces and costly classical diagonalization. We introduce QSCI-RBM, a compact subspace generation protocol based on Restricted Boltzmann Machines (RBMs), and integrate it within Density Matrix Embedding Theory (DMET). The RBM learns the distribution of dominant determinants from hardware samples, generating high-probability configurations. Sampling was performed on IBM Heron R1 (\texttt{ibm\_torino}) for CH5NO, R2 (\texttt{ibm\_fez}) for C2H4, and R3 (\texttt{ibm\_boston}) for the larger H2O benchmark and the protein-ligand system. After validating on these benchmarks of increasing active-space size, we simulate the inhibitor Carmofur bound to the SARS-CoV-2 main protease (M\textsuperscript{pro}), fragmented into ten DMET impurities. Both DMET-QSCI-RBM and a DMET-SQD baseline reach chemical-potential self-consistency within four μ-iterations. At the converged chemical potential, DMET-QSCI-RBM lies within 0.05 mHa of DMET-FCI while accessing on average only 0.22% of the symmetry-preserving configuration space, versus 0.55 mHa at 33.0% for DMET-SQD, leading to a ∼153-fold subspace reduction. RBM-assisted expansion thus yields far more compact hardware-derived determinant spaces at equal or better accuracy, reducing classical overhead for scalable quantum embedding of biological systems on real hardware.
Related
- Generation of membrane-permeable cyclic peptides inhibiting protein–protein interaction
- Establishing standardized evaluation protocols for peptide-target interaction prediction: the PTI-TAPE consensus framework eDelphi
- A novel matrix multiplication framework for modeling genotype-by-environment interaction in genomic prediction
- Refining the TCGA Framework: The Pivotal Role of Targeted Next-Generation Sequencing in Resolving Diagnostic Ambiguities in Endometrial Cancer—Systematic Review
- AIPFusion-Net: an interaction-fusion framework for anti-inflammatory peptide prediction using protein language model embeddings, local sequence patterns, and handcrafted descriptors.
BibTeX
@misc{patra2026machinelearned,
title = {{Machine-Learned Compact Subspace Generation for Quantum Selected Configuration Interaction within Density Matrix Embedding Framework}},
author = {Ashish Kumar Patra and Anurag K. S. V. and Ruchika Bhat and Sai Shankar P. and Rahul Maitra and Jaiganesh G},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2607.20585},
eprint = {2607.20585},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2607.20585},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Machine-Learned Compact Subspace Generation for Quantum Selected Configuration Interaction within Density Matrix Embedding Framework
AU - Ashish Kumar Patra
AU - Anurag K. S. V.
AU - Ruchika Bhat
AU - Sai Shankar P.
AU - Rahul Maitra
AU - Jaiganesh G
PY - 2026
DA - 2026/10/31
PB - arXiv
DO - 10.48550/arxiv.2607.20585
UR - https://arxiv.org/abs/2607.20585
AB - Sample-based quantum diagonalization (SQD), an extension of quantum selected configuration interaction (QSCI), is a promising hybrid quantum-classical paradigm for molecular ground-state energies. By replacing variational optimization with quantum sampling, QSCI avoids barren plateaus and directly reconstructs correlated wavefunctions. However, existing configuration recovery mainly enforces symmetry constraints without targeting the most physically relevant configurations, often yielding oversized subspaces and costly classical diagonalization. We introduce QSCI-RBM, a compact subspace generation protocol based on Restricted Boltzmann Machines (RBMs), and integrate it within Density Matrix Embedding Theory (DMET). The RBM learns the distribution of dominant determinants from hardware samples, generating high-probability configurations. Sampling was performed on IBM Heron R1 (\texttt{ibm\_torino}) for CH_5NO, R2 (\texttt{ibm\_fez}) for C_2H_4, and R3 (\texttt{ibm\_boston}) for the larger H_2O benchmark and the protein-ligand system. After validating on these benchmarks of increasing active-space size, we simulate the inhibitor Carmofur bound to the SARS-CoV-2 main protease (M\textsuperscript{pro}), fragmented into ten DMET impurities. Both DMET-QSCI-RBM and a DMET-SQD baseline reach chemical-potential self-consistency within four μ-iterations. At the converged chemical potential, DMET-QSCI-RBM lies within 0.05 mHa of DMET-FCI while accessing on average only 0.22% of the symmetry-preserving configuration space, versus 0.55 mHa at 33.0% for DMET-SQD, leading to a ∼153-fold subspace reduction. RBM-assisted expansion thus yields far more compact hardware-derived determinant spaces at equal or better accuracy, reducing classical overhead for scalable quantum embedding of biological systems on real hardware.
ER -