Coupled Cluster con M\=oLe: Molecular Orbital Learning for Neural Wavefunctions

2026-02-23

AI Summary

This work proposes MōLe, an equivariant machine learning architecture that addresses the high computational cost of traditional coupled-cluster (CC) methods, which, despite their accuracy, struggle to scale to large molecular systems. MōLe is the first model to enable end-to-end prediction of CC excitation amplitudes directly from Hartree–Fock molecular orbitals. By integrating quantum chemical representations with equivariant neural networks, the method achieves remarkable data efficiency and strong generalization—accurately predicting amplitudes for larger molecules and non-equilibrium geometries despite being trained exclusively on equilibrium configurations of small molecules. Experimental results demonstrate that MōLe substantially reduces the number of iterations required in CC calculations, thereby accelerating high-accuracy quantum chemical simulations.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search

Application Category

• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

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 (M\=oLe) 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 geometries, despite being trained only on small equilibrium geometries. Finally, we also examine its ability to reduce the number of cycles required to converge CC calculations. M\=oLe can set the foundations for high-accuracy wavefunction-based ML architectures to accelerate molecular design and complement force-field approaches.

Problem

• Coupled Cluster
• computational cost
• molecular properties
• quantum chemistry
• accuracy

Innovation

• Coupled Cluster
• Machine Learning
• Equivariant Neural Networks
• Excitation Amplitudes
• Molecular Orbital Learning