Machine Learning the Complete Active-Space Wavefunction Reconstruction
Abstract
Machine-learned interatomic potentials enable multireference-quality reaction dynamics but do not retain the electronic structure underlying the predicted energies and forces. Recovering this information by independently performing complete active space self-consistent field (CASSCF) calculations is both expensive and potentially inconsistent. We introduce the Complete Active-Space Wave Function Reconstruction (CAS-WR) model, which learns the geometry-dependent electronic structure from CASSCF data and reconstructs the CAS wave function without repeated orbital optimization. Combined with uncertainty-based active learning, CAS-WR is demonstrated for the Diels--Alder reaction and \ce{TiC+}-catalyzed C--H activation. The approach enables trajectory-resolved analysis of multiconfigurational electronic structure across reactive ensembles. Preliminary applications to excited-state proton transfer and transition-metal photocatalysis suggest that CAS-WR can provide a general route toward recovering electronic structure along complex ground- and excited-state reactive dynamics.
Related
- Smart Machine Learning for Solving the Full Electron-Nuclear Schrödinger Equation with a Quantum-Chemically Inspired Wavefunction Ansatz
- Ensuring Complete Consistency between Hops and Wavefunction Propagation in Fewest-Switches Surface Hopping
- Depth-aware machine learning for 3D reconstruction of fibrous porous electrodes by FIB-SEM tomography
- MLIP Detective: Active Failure Mode Discovery Beyond Benchmark Scores for Machine-Learning Interatomic Potentials
- Data-Efficient Machine Learning Potentials for Organic Reactions through Active-Learning-Guided Fine-Tuning
BibTeX
@misc{seal2026machine,
title = {{Machine Learning the Complete Active-Space Wavefunction Reconstruction}},
author = {Aniruddha Seal and Bhavnesh Jangid and Helen S. Clifford and Andrew L. Ferguson and Laura Gagliardi},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.06779},
eprint = {2611.06779},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.06779},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Machine Learning the Complete Active-Space Wavefunction Reconstruction
AU - Aniruddha Seal
AU - Bhavnesh Jangid
AU - Helen S. Clifford
AU - Andrew L. Ferguson
AU - Laura Gagliardi
PY - 2026
DA - 2026/11/04
PB - arXiv
DO - 10.48550/arxiv.2611.06779
UR - https://arxiv.org/abs/2611.06779
AB - Machine-learned interatomic potentials enable multireference-quality reaction dynamics but do not retain the electronic structure underlying the predicted energies and forces. Recovering this information by independently performing complete active space self-consistent field (CASSCF) calculations is both expensive and potentially inconsistent. We introduce the Complete Active-Space Wave Function Reconstruction (CAS-WR) model, which learns the geometry-dependent electronic structure from CASSCF data and reconstructs the CAS wave function without repeated orbital optimization. Combined with uncertainty-based active learning, CAS-WR is demonstrated for the Diels--Alder reaction and \ce{TiC+}-catalyzed C--H activation. The approach enables trajectory-resolved analysis of multiconfigurational electronic structure across reactive ensembles. Preliminary applications to excited-state proton transfer and transition-metal photocatalysis suggest that CAS-WR can provide a general route toward recovering electronic structure along complex ground- and excited-state reactive dynamics.
ER -