Electronic Spectra from TDDFT and Machine Learning in Chemical Space

Abstract

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 singles and doubles (CC2) spectra from TDDFT counterparts, or even from DFT gap. We applied this approach to low-lying singlet-singlet vertical electronic spectra of over 20 thousand 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 thousand molecules, CC2 excitation energies can be reproduced to within $\pm$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.

References in corpus (5)

• Kernel density estimation via diffusion
• Big Data meets Quantum Chemistry Approximations: The $\Delta$-Machine Learning Approach
• Machine learning for many-body physics: The case of the Anderson impurity model
• Recognizing molecular patterns by machine learning: an agnostic structural definition of the hydrogen bond
• Using molecular similarity to construct accurate semiempirical electron structure theories

Cited by in corpus (36)

• DScribe: Library of Descriptors for Machine Learning in Materials Science
• ChemRL-GEM: Geometry Enhanced Molecular Representation Learning for Property Prediction
• Machine learning for electronically excited states of molecules
• Prediction model of band-gap for AX binary compounds by combination of density functional theory calculations and machine learning techniques
• Autonomous discovery in the chemical sciences part II: Outlook
• Machine Learning for Quantum Mechanical Properties of Atoms in Molecules
• Perspective on integrating machine learning into computational chemistry and materials science
• Structure-based Sampling and Self-correcting Machine Learning for Accurate Calculations of Potential Energy Surfaces and Vibrational Levels
• Machine Learning Exciton Dynamics
• A Deep Dive into Machine Learning Density Functional Theory for Materials Science and Chemistry
• Solid Harmonic Wavelet Scattering for Predictions of Molecule Properties
• PubChemQC PM6: A dataset of 221 million molecules with optimized molecular geometries and electronic properties
• Chemical diversity in molecular orbital energy predictions with kernel ridge regression
• Deep Learning for UV Absorption Spectra with SchNarc: First Steps Towards Transferability in Chemical Compound Space
• Regression-clustering for Improved Accuracy and Training Cost with Molecular-Orbital-Based Machine Learning
• Olympus: a benchmarking framework for noisy optimization and experiment planning
• Physically inspired deep learning of molecular excitations and photoemission spectra
• The quest to simulate excited-state dynamics of transition metal complexes
• The role of tachysterol in vitamin D photosynthesis - A non-adiabatic molecular dynamics study
• Accurate Molecular-Orbital-Based Machine Learning Energies via Unsupervised Clustering of Chemical Space
• The Chemical Space of B, N-substituted Polycyclic Aromatic Hydrocarbons: Combinatorial Enumeration and High-Throughput First-Principles Modeling
• UV-Visible Absorption Spectra of Solvated Molecules by Quantum Chemical Machine Learning
• Predicting excited states from ground state wavefunction by supervised quantum machine learning
• Quantum chemical roots of machine-learning molecular similarity descriptors
• Molecular Quantum Chemical Data Sets and Databases for Machine Learning Potentials
• The fundamentals of quantum machine learning
• A Snapshot of Time-Dependent Density-Functional Theory
• Generalizable, Fast, and Accurate DeepQSPR with fastprop
• Machine learning at the atomic-scale
• Two-dimensional electronic spectroscopy in the condensed phase using equivariant transformer accelerated molecular dynamics simulations
• Data-driven approach for benchmarking DFTB-approximate excited state methods
• Electronic Descriptors for Supervised Spectroscopic Predictions
• Thawed Gaussian wavepacket dynamics with $\Delta$-machine learned potentials
• Excited-State Response Theory Within the Context of the Coupled-Cluster Formalism
• Data-Driven Modeling of S0 -> S1 Excitation Energy in the BODIPY Chemical Space: High-Throughput Computation, Quantum Machine Learning, and Inverse Design
• Integer linear programming for unsupervised training set selection in molecular machine learning