Multi-task learning for molecular electronic structure approaching coupled-cluster accuracy

Machine learning (ML) plays an important role in quantum chemistry, providing fast-to-evaluate predictive models for various properties of molecules. However, as most existing ML models for molecular electronic properties use density function theory (DFT) databases as the ground truth in training, their prediction accuracy cannot go beyond the DFT. In this work, we developed a unified ML method for electronic structures of organic molecules using the gold-standard CCSD(T) calculations as training data. Tested on hydrocarbon molecules, our model outperforms the DFT with the widely-used B3LYP functional in both computation costs and prediction accuracy of various quantum chemical properties. We apply the model to aromatic compounds and semiconducting polymers on both ground state and excited state properties, demonstrating its accuracy and generalization capability to complex systems that are hard to calculate using CCSD(T)-level methods.