Quantum-chemical insights from deep tensor neural networks.
Schütt KT, Arbabzadah F, Chmiela S, Müller KR, Tkatchenko A. · 2017
Read Quantum-chemical insights from deep tensor neural networks. by Schütt KT, Arbabzadah F, Chmiela S, Müller KR, Tkatchenko A. (2017) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc5228054.txt.
Introduction
Kristof T. Schütt, Farhad Arbabzadah, Stefan Chmiela, Klaus-Robert Müller, and Alexandre Tkatchenko built deep tensor neural networks for quantum chemistry in Nature Communications in 2017. Machine learning had reshaped fields from image search to speech, and the team asked whether it could also reveal how quantum mechanics shapes the behavior of molecules, instead of only crunching numbers.
Their model merges ideas from many-body physics with a neural network built for the task. It predicts molecular properties across chemical space with errors near one kilocalorie per mole, the accuracy chemists aim for. Because its readouts resolve position and element, it can estimate atomic energies and local chemical potentials inside a molecule. As a test of insight, the model grouped aromatic rings by stability, and it also handled isomer energies and molecules with unusual electronic structure well.
Catalog record
| CREATOR | Schütt KT, Arbabzadah F, Chmiela S, Müller KR, Tkatchenko A. |
|---|---|
| YEAR | 2017 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 39 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Quantum-chemical insights from deep tensor neural networks By Kristof T Schütt, Farhad Arbabzadah, Stefan Chmiela, Klaus R Müller, Alexandre Tkatchenko Published in: Nature Communications DOI: 10.1038/ncomms13890 ABSTRACT Abstract Learning from data has led to paradigm shifts in a multitude of disciplines, including web, text and image search, speech recognition, as well as bioinformatics. Can machine learning enable similar breakthroughs in understanding quantum many-body systems? Here we develop an efficient deep learning approach that enables spatially and chemically resolved insights into quantum-mechanical observables of molecular systems. We unify concepts from many-body Hamiltonians with purpose-designed deep tensor neural networks, which leads to size-extensive and uniformly accurate (1 kcal mol−1) predictions in compositional and configurational chemical space for molecules of intermediate size. As an example of chemical relevance, the model reveals a classification of aromatic rings with respect to their stability. Further applications of our model for predicting atomic energies and local chemical potentials in molecules, reliable isomer energies, and molecules with peculiar electronic structure demonstrate the potential of machine learning for revealing insights into complex quantum-chemical systems. Chemistry permeates all aspects of our life, from the development of new drugs to the food that we consume and materials we use on a daily basis. Chemists rely on empirical observations based on creative and painstaking experimentation that leads to eventual discoveries of molecules and materials with desired properties and mechanisms to synthesize them. Many discoveries in chemistry can be guided by searching large databases of experimental or computational molecular structures and properties by using concepts based on chemical similarity. Because the structure and properties of molecules are determined by the laws of quantum mechanics, ultimately chemical discovery must be based on fundamental quantum principles. Indeed, electronic structure calculations and intelligent data analysis (machine learning) have recently been combined aiming towards the goal of accelerated discovery of chemicals with desired properties1,2,3,4,5,6,7,8. However, so far the majority of these pioneering efforts have focused on the construction of reduced models trained on large data sets of density-functional theory calculations. In this work, we develop an efficient deep learning approach that enables spatially and chemically resolved insights into quantum-mechanical properties of molecular systems beyond those trivially contained in the training dataset. Obviously, computational models are not predictive if they lack accuracy. In addition to being interpretable, size-extensive and efficient, our deep tensor neural network (DTNN) approach is uniformly accurate (1 kcal mol−1) throughout compositional and configurational chemical space. On the more fundamental side, the mathematical construction of the DTNN model provides statistically rigorous partitioning of extensive molecular properties into atomic contributions—a long-standing challenge for quantum-mechanical calculations of molecules. RESULTS Molecular deep tensor neural networks It is common to use a carefully chosen representation of the problem at hand as a basis for machine learning9,10,11. For example, molecules can be represented as Coulomb matrices7,12,13, scattering transforms14, bags of bonds15, smooth overlap of atomic positions16,17 or generalized symmetry functions18,19. Kernel-based learning of molecular properties transforms these representations non-linearly by virtue of kernel functions. In contrast, deep neural networks20 are able to infer the underlying regularities and learn an efficient representation in a layer-wise fashion21. Molecular properties are governed by the laws of quantum mechanics, which yield the remarkable flexibility of chemical systems, but also impose constraints on the behaviour of bonding in molecules. The approach presented here utilizes the many-body Hamiltonian concept for the construction of the DTNN architecture (Fig. 1), embracing the principles of quantum chemistry, while maintaining the full flexibility of a complex data-driven learning machine. [Figure: Prediction and explanation of molecular energies with a deep tensor neural network.(a) Molecules are encoded as input for the neural network by a vector of nuclear charges and an inter-atomic distance matrix. This description is complete and invariant to rotation and translation. (b) Illustration of the network architecture. Each atom type corresponds to a vector of coefficients , which is repeatedly refined by interactions vij. The interactions depend on the current representation , as well as the distance Dij to an atom j. After T iterations, an energy contribution Ei is predicted for the final coefficient vector . The molecular energy E is the sum over these atomic contributions. (c) Mean absolute errors of predictions for the GDB-9 dataset of 133,885 molecules as a function of the number of atoms. The employed neural network uses two interaction passes (T=2) and 50,000 reference calculation during training. The inset shows the error of an equivalent network trained on 5,000 GDB-9 molecules with 20 or more atoms, as small molecules with 15 or less atoms are added to the training set. (d) Extract from the calculated (black) and predicted (orange) molecular dynamics trajectory of toluene. The curve on the right shows the agreement of the predicted and calculated energy distributions. (e) Energy contribution Eprobe (or local chemical potential , see text) of a hydrogen test charge on a isosurface for various molecules from the GDB-9 dataset for a DTNN model with T=2.] DTNN receives molecular structures through a vector of nuclear charges Z and a matrix of atomic distances D ensuring rotational and translational invariance by construction (Fig. 1a). The distances are expanded in a Gaussian basis, yielding a feature vector , which accounts for the different nature of interactions at various distance regimes. Similar approaches have been applied to the entries of the Coulomb matrix for the prediction of molecular properties before12. The total energy EM for the molecule M composed of N atoms is written as a sum over N atomic energy contributions Ei, thus satisfying permutational invariance with respect to atom indexing. Each atom i is represented by a coefficient vector , where B is the number of basis functions, or features. Motivated by quantum-chemical atomic basis set expansions, we assign an atom type-specific descriptor vector to these coefficients . Subsequently, this atomic expansion is repeatedly refined by pairwise interactions with the surrounding atoms where the interaction term vij reflects the influence of atom j at a distance Dij on atom i. Note that this refinement step is seamlessly integrated into the architecture of the molecular DTNN, and is therefore adapted throughout the learning process. In Supplementary Discussion, we show the relation to convolutional neural networks that have been applied to images, speech and text with great success because of their ability to capture local structure22,23,24,25,26,27. Considering a molecule as a graph, T refinements of the coefficient vectors are comprised of all walks of length T through the molecule ending at the corresponding atom28,29. From the point of view of many-body interatomic interactions, subsequent refinement steps t correlate atomic neighbourhoods with increasing complexity. While the initial atomic representations only consider isolated atoms, the interaction terms characterize how the basis functions of two atoms overlap with each other at a certain distance. Each refinement step is supposed to reduce these overlaps, thereby embedding the atoms of the molecule into their chemical environment. Following this procedure, the DTNN implicitly learns an atom-centered basis that is unique and efficient with respect to the property to be predicted. Non-linear coupling between the atomic vector features and the interatomic distances is achieved by a tensor layer30,31,32, such that the coefficient k of the refinement is given by where bk is the bias of feature k