Quantum evolution kernel for molecular machine learning on programmable neutral-atom arrays
6 affiliations
- 1Department of Chemistry, University of Calgary, 2500 University Drive NW, Calgary, Alberta, Canada, T2N 1N4
- 2Institute for Quantum Science and Technology, University of Calgary, 2500 University Drive NW, Calgary, Alberta, Canada, T2N 1N4
- 3Faculty of Graduate Studies, University of Calgary, 2500 University Drive NW, Calgary, AB, Canada, T2N 1N4
- 4Département de génie électrique et de génie informatique, Université de Sherbrooke, 2500, boulevard de l’Université, Sherbrooke, Québec, Canada, J1K 2R1
- 5Institut quantique, 2500 Boulevard de l’Université, Sherbrooke, Québec, Canada, J1K 2R1
- 6Centre for Molecular Simulation, University of Calgary, 2500 University Drive NW, Calgary, Alberta, Canada, T2N 1N4
Abstract
Neutral-atom quantum processors can embed molecular graphs natively through the geometry of atom arrays, which makes them a natural platform for graph-based quantum kernels. Existing quantum evolution kernels, however, encode limited chemical information and have been applied to molecular classification but not to regression. Here we introduce ChemQEK, a quantum evolution kernel framework for molecular property prediction on analog neutral-atom quantum processors. ChemQEK encodes atom-specific descriptors through local addressability and atom-pair London dispersion coefficients (Cij6) into the parameters of the Rydberg Hamiltonian, and it combines global, local and pairwise observables of the evolved state within a single kernel. We further introduce kernel pooling strategies that combine information across evolution times and addressability schemes. Using noiseless emulation, we evaluate ChemQEK with kernel-based regression models on the ESOL, FreeSolv, Jean-Claude Bradley Open Melting Point, and QM9 datasets. Pooling over both evolution times and addressability schemes outperforms pooling over either alone and yields test R2 values of 0.90, 0.84, and 0.69 for ESOL, FreeSolv, and melting point, respectively. On QM9, ChemQEK outperforms the quantum evolution kernel and generalized-distance quantum-correlation frameworks, each of which relies on a single observable type, for all eight properties considered. The QM9 experiments also reveal property dependent trade-offs between predictive performance and the number of quantum feature configurations, with up to ∼79\% fewer measurements retaining competitive performance for several properties. These results highlight the importance of chemically informed encoding, complementary observables, and multidimensional kernel pooling for quantum kernel-based molecular property prediction.
Related
- A Machine Learning Approach to logBB Prediction Combining Molecular Docking and Quantum-Chemical Descriptors
- Cost-Effective Quantum-Mechanical Workflows for Molecular Machine Learning
- Developing a Quantum Physics-Inspired Machine Learning Model for Partition Coefficient Prediction and Comparing Its Performance with Empirical Descriptors and Molecular Graph-Based Models
- Accuracy, Reliability, and Efficiency of a MACE Machine-Learning Interatomic Potential for Aspirin: A Reproducible Benchmark for Molecular Simulation
- Embedded Morgan Fingerprints for more efficient molecular property predictions with machine learning
BibTeX
@misc{sakhuja2026quantum,
title = {{Quantum evolution kernel for molecular machine learning on programmable neutral-atom arrays}},
author = {Saiyam Sakhuja and Victor Drouin-Touchette and Viki Kumar Prasad},
howpublished = {ChemRxiv},
year = {2026},
month = oct,
doi = {10.26434/chemrxiv.15011111/v1},
url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15011111/v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Quantum evolution kernel for molecular machine learning on programmable neutral-atom arrays
AU - Saiyam Sakhuja
AU - Victor Drouin-Touchette
AU - Viki Kumar Prasad
PY - 2026
DA - 2026/10/27
PB - ChemRxiv
DO - 10.26434/chemrxiv.15011111/v1
UR - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15011111/v1
AB - Neutral-atom quantum processors can embed molecular graphs natively through the geometry of atom arrays, which makes them a natural platform for graph-based quantum kernels. Existing quantum evolution kernels, however, encode limited chemical information and have been applied to molecular classification but not to regression. Here we introduce ChemQEK, a quantum evolution kernel framework for molecular property prediction on analog neutral-atom quantum processors. ChemQEK encodes atom-specific descriptors through local addressability and atom-pair London dispersion coefficients (C^ij_6) into the parameters of the Rydberg Hamiltonian, and it combines global, local and pairwise observables of the evolved state within a single kernel. We further introduce kernel pooling strategies that combine information across evolution times and addressability schemes. Using noiseless emulation, we evaluate ChemQEK with kernel-based regression models on the ESOL, FreeSolv, Jean-Claude Bradley Open Melting Point, and QM9 datasets. Pooling over both evolution times and addressability schemes outperforms pooling over either alone and yields test R^2 values of 0.90, 0.84, and 0.69 for ESOL, FreeSolv, and melting point, respectively. On QM9, ChemQEK outperforms the quantum evolution kernel and generalized-distance quantum-correlation frameworks, each of which relies on a single observable type, for all eight properties considered. The QM9 experiments also reveal property dependent trade-offs between predictive performance and the number of quantum feature configurations, with up to ∼79\% fewer measurements retaining competitive performance for several properties. These results highlight the importance of chemically informed encoding, complementary observables, and multidimensional kernel pooling for quantum kernel-based molecular property prediction.
ER -