Electronic spectra from TDDFT and machine learning in chemical space.

By R. Ramakrishnan, M. Hartmann, E. Tapavicza, O. V. Lilienfeld. 2015

287 citations recorded by the source corpus.

Due to its favorable computational efficiency, time-dependent (TD) density functional theory (DFT) enables the prediction of electronic spectra in a high-throughput manner across chemical space. Its predictions, however, can be quite inaccurate. We resolve this issue with machine learning models trained on deviations of reference second-order approximate coupled-cluster (CC2) singles and doubles spectra from TDDFT counterparts, or even from DFT gap. We applied this approach to low-lying singlet-singlet vertical electronic spectra of over 20 000 synthetically feasible small organic molecules with up to eight CONF atoms. The prediction errors decay monotonously as a function of training set size. For a training set of 10 000 molecules, CC2 excitation energies can be reproduced to within ±0.1 eV for the remaining molecules. Analysis of our spectral database via chromophore counting suggests that even higher accuracies can be achieved. Based on the evidence collected, we discuss open challenges associated with data-driven modeling of high-lying spectra and transition intensities.

Subfields: Atomic and Molecular Physics, and Optics, Materials Chemistry, Spectroscopy

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Atomic and Molecular Physics, and Optics

In Materials Chemistry

In Spectroscopy

Citations in the corpus

PotentialNet for Molecular Property Prediction (2018, methodology): “We utilize MAE for consistency with the original proposal of the database 48 .”
Simplifying Graph Convolutional Networks (2019, methodology): “Similarly, on QM8 quantum chemistry dataset (Ramakrishnan et al., 2015), more advanced AdaLNet and LNet (Liao et al.”
Geometry-enhanced molecular representation learning for property prediction (2021, methodology): “With respect to the regression datasets, we use root mean square error (RMSE) for FreeSolv37, ESOL38 and Lipo39, whereas we use mean average error (MAE) for QM740, QM841 and QM942.”
Machine learning exciton dynamics (2015, unknown): “[41] Ramakrishnan, R, Hartmann, M, Tapavicza, E, & von Lilienfeld, O.”
Hyperbolic Graph Neural Networks (2019, unknown): “However, see the appendix for results on QM8 [40, 39] and QM9 [40, 38].”
Do Large Scale Molecular Language Representations Capture Important Structural Information? (2021, methodology): “QM8 Ruddigkeit et al. (2012); Ramakrishnan et al. (2015) consists of four excited state properties, each calculated by three different quantum mechanical methods, on 22k samples comprised of up to eight heavy atoms.”
LanczosNet: Multi-Scale Deep Graph Convolutional Networks (2019, methodology): “…[39] 13.19 ± 0.04 12.95 ± 0.11 GPNN [40] 12.81 ± 0.80 12.39 ± 0.77 GAT [33] 11 We then benchmark all algorithms on QM8 quantum chemistry dataset which comes from a recent study on modeling quantum mechanical calculations of electronic spectra and excited state energy of small molecules [64].”
Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development (2021, methodology): “[115] Raghunathan Ramakrishnan, Mia Hartmann, Enrico Tapavicza, and O Anatole Von Lilienfeld.”