![[2602.20232] Coupled Cluster con MōLe: Molecular Orbital Learning for Neural Wavefunctions](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.20232] Coupled Cluster con MōLe: Molecular Orbital Learning for Neural Wavefunctions
Summary
The paper introduces MōLe, a machine learning architecture that predicts excitation amplitudes in coupled-cluster theory, enhancing molecular property calculations with improved efficiency and accuracy.
Why It Matters
This research addresses the limitations of traditional density functional theory (DFT) in molecular property predictions by leveraging machine learning to enhance coupled-cluster methods, potentially accelerating molecular design and improving computational efficiency in quantum chemistry.
Key Takeaways
- MōLe predicts excitation amplitudes from Hartree-Fock molecular orbitals, enhancing coupled-cluster calculations.
- The model demonstrates high data efficiency and generalization to larger molecules despite limited training data.
- MōLe can reduce convergence cycles in coupled-cluster calculations, improving computational efficiency.
Computer Science > Machine Learning arXiv:2602.20232 (cs) [Submitted on 23 Feb 2026] Title:Coupled Cluster con MōLe: Molecular Orbital Learning for Neural Wavefunctions Authors:Luca Thiede, Abdulrahman Aldossary, Andreas Burger, Jorge Arturo Campos-Gonzalez-Angulo, Ning Wang, Alexander Zook, Melisa Alkan, Kouhei Nakaji, Taylor Lee Patti, Jérôme Florian Gonthier, Mohammad Ghazi Vakili, Alán Aspuru-Guzik View a PDF of the paper titled Coupled Cluster con M\=oLe: Molecular Orbital Learning for Neural Wavefunctions, by Luca Thiede and 11 other authors View PDF HTML (experimental) Abstract:Density functional theory (DFT) is the most widely used method for calculating molecular properties; however, its accuracy is often insufficient for quantitative predictions. Coupled-cluster (CC) theory is the most successful method for achieving accuracy beyond DFT and for predicting properties that closely align with experiment. It is known as the ''gold standard'' of quantum chemistry. Unfortunately, the high computational cost of CC limits its widespread applicability. In this work, we present the Molecular Orbital Learning (MōLe) architecture, an equivariant machine learning model that directly predicts CC's core mathematical objects, the excitation amplitudes, from the mean-field Hartree-Fock molecular orbitals as inputs. We test various aspects of our model and demonstrate its remarkable data efficiency and out-of-distribution generalization to larger molecules and off-equilibrium geom...
