Quantum-Chemical Insights from Interpretable Atomistic Neural Networks
| dc.contributor.author | Schütt, Kristof T. | |
| dc.contributor.author | Gastegger, Michael | |
| dc.contributor.author | Tkatchenko, Alexandre | |
| dc.contributor.author | Müller, Klaus-Robert | |
| dc.date.accessioned | 2020-06-19T08:06:56Z | |
| dc.date.available | 2020-06-19T08:06:56Z | |
| dc.date.issued | 2019-09-10 | |
| dc.description.abstract | With the rise of deep neural networks for quantum chemistry applications, there is a pressing need for architectures that, beyond delivering accurate predictions of chemical properties, are readily interpretable by researchers. Here, we describe interpretation techniques for atomistic neural networks on the example of Behler–Parrinello networks as well as the end-to-end model SchNet. Both models obtain predictions of chemical properties by aggregating atom-wise contributions. These latent variables can serve as local explanations of a prediction and are obtained during training without additional cost. Due to their correspondence to well-known chemical concepts such as atomic energies and partial charges, these atom-wise explanations enable insights not only about the model but more importantly about the underlying quantum-chemical regularities. We generalize from atomistic explanations to 3d space, thus obtaining spatially resolved visualizations which further improve interpretability. Finally, we analyze learned embeddings of chemical elements that exhibit a partial ordering that resembles the order of the periodic table. As the examined neural networks show excellent agreement with chemical knowledge, the presented techniques open up new venues for data-driven research in chemistry, physics and materials science. | en |
| dc.description.sponsorship | EC/H2020/792572/EU/Machine Learning for Catalytic Carbon Dioxide Activation/MachineCat | en |
| dc.description.sponsorship | BMBF, 01IS14013A, Verbundprojekt: BBDC - Berliner Kompetenzzentrum für Big Data | en |
| dc.description.sponsorship | BMBF, 01IS18037A, Verbundprojekt BIFOLD-BZML: Berlin Institute for the Foundations of Learning and Data | en |
| dc.description.sponsorship | EC/H2020/725291/EU/Beyond Static Molecules: Modeling Quantum Fluctuations in Complex Molecular Environments/BeStMo | en |
| dc.identifier.eissn | 1611-3349 | |
| dc.identifier.isbn | 978-3-030-28954-6 | |
| dc.identifier.isbn | 978-3-030-28953-9 | |
| dc.identifier.issn | 0302-9743 | |
| dc.identifier.uri | https://depositonce.tu-berlin.de/handle/11303/11437 | |
| dc.identifier.uri | http://dx.doi.org/10.14279/depositonce-10318 | |
| dc.language.iso | en | en |
| dc.rights.uri | http://rightsstatements.org/vocab/InC/1.0/ | en |
| dc.subject.ddc | 541 Physikalische Chemie | de |
| dc.subject.ddc | 006 Spezielle Computerverfahren | de |
| dc.subject.other | machine learning | en |
| dc.subject.other | chemistry | en |
| dc.subject.other | explanation | en |
| dc.title | Quantum-Chemical Insights from Interpretable Atomistic Neural Networks | en |
| dc.type | Book Part | en |
| dc.type.version | acceptedVersion | en |
| dcterms.bibliographicCitation.booktitle | Explainable AI: Interpreting, Explaining and Visualizing Deep Learning | en |
| dcterms.bibliographicCitation.doi | 10.1007/978-3-030-28954-6_17 | en |
| dcterms.bibliographicCitation.editor | Samek, Wojciech | |
| dcterms.bibliographicCitation.editor | Montavon, Grégoire | |
| dcterms.bibliographicCitation.editor | Vedaldi, Andrea | |
| dcterms.bibliographicCitation.editor | Hansen, Lars Kai | |
| dcterms.bibliographicCitation.editor | Müller, Klaus-Robert | |
| dcterms.bibliographicCitation.originalpublishername | Springer | en |
| dcterms.bibliographicCitation.originalpublisherplace | Cham | en |
| dcterms.bibliographicCitation.pageend | 330 | en |
| dcterms.bibliographicCitation.pagestart | 311 | en |
| tub.accessrights.dnb | free | en |
| tub.affiliation | Fak. 4 Elektrotechnik und Informatik::Inst. Softwaretechnik und Theoretische Informatik::FG Maschinelles Lernen | de |
| tub.affiliation.faculty | Fak. 4 Elektrotechnik und Informatik | de |
| tub.affiliation.group | FG Maschinelles Lernen | de |
| tub.affiliation.institute | Inst. Softwaretechnik und Theoretische Informatik | de |
| tub.publisher.universityorinstitution | Technische Universität Berlin | en |
| tub.series.issuenumber | 11700 | en |
| tub.series.name | Lecture Notes in Computer Science | en |