Prediction of Atomization Energy Using Graph Kernel and Active Learning

Abstract

Data-driven prediction of molecular properties presents unique challenges to the design of machine learning methods concerning data structure/dimensionality, symmetry adaption, and confidence management. In this paper, we present a kernel-based pipeline that can learn and predict the atomization energy of molecules with high accuracy. The framework employs Gaussian process regression to perform predictions based on the similarity between molecules, which is computed using the marginalized graph kernel. To apply the marginalized graph kernel, a spatial adjacency rule is first employed to convert molecules into graphs whose vertices and edges are labeled by elements and interatomic distances, respectively. We then derive formulas for the efficient evaluation of the kernel. Specific functional components for the marginalized graph kernel are proposed, while the effect of the associated hyperparameters on accuracy and predictive confidence are examined. We show that the graph kernel is particularly suitable for predicting extensive properties because its convolutional structure coincides with that of the covariance formula between sums of random variables. Using an active learning procedure, we demonstrate that the proposed method can achieve a mean absolute error of 0.62 +- 0.01 kcal/mol using as few as 2000 training samples on the QM7 data set.

References in corpus (13)

• On representing chemical environments
• Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
• ANI-1: An extensible neural network potential with DFT accuracy at force field computational cost
• Quantum-Chemical Insights from Deep Tensor Neural Networks
• Comparing molecules and solids across structural and alchemical space
• Machine learning prediction errors better than DFT accuracy
• Unified Representation of Molecules and Crystals for Machine Learning
• Understanding molecular representations in machine learning: The role of uniqueness and target similarity
• Machine learning of molecular properties: locality and active learning
• Learning molecular energies using localized graph kernels
• Active learning of constitutive relation from mesoscopic dynamics for macroscopic modeling of non-Newtonian flows
• A Projected Preconditioned Conjugate Gradient Algorithm for Computing Many Extreme Eigenpairs of a Hermitian Matrix
• An Atomistic Fingerprint Algorithm for Learning Ab Initio Molecular Force Fields

Cited by in corpus (10)

• The Open Catalyst 2020 (OC20) Dataset and Community Challenges
• Predicting binding motifs of complex adsorbates using machine learning with a physics-inspired graph representation
• Graphical Gaussian Process Regression Model for Aqueous Solvation Free Energy Prediction of Organic Molecules in Redox Flow Battery
• A High-Throughput Solver for Marginalized Graph Kernels on GPU
• Hybrid localized graph kernel for machine learning energy-related properties of molecules and solids
• Designing compact training sets for data-driven molecular property prediction
• Detecting Label Noise via Leave-One-Out Cross-Validation
• Incorporating electronic information into Machine Learning potential energy surfaces via approaching the ground-state electronic energy as a function of atom-based electronic populations
• Nonlinear Matrix Approximation with Radial Basis Function Components
• Optimal Sampling Density for Nonparametric Regression