Inc 2 - Machine Learning
Group Leaders
These observations suggest that ML methods can now be adapted to tackle the more generic challenge of materials discovery and design. In order to resolve the problem of extrapolation versus interpolation we plan to exploit that training and application of ML models is "agnostic" in the sense that origin and choice of the training data is irrelevant for formal construction of the ML model, due to its inherently inductive nature. As such, ML models can be improved or updated on a regular basis, through adaptation of training set size and composition, compound's representations, or locally adaptive regression functions. More specifically within this incubator project Inc 2 — Active Machine Learning for Computational Materials Design — we investigate possible ways to systematically accelerate the computational identification of promising materials candidates through combinations of supervised and unsupervised ML models, property optimization algorithms, and active learning.
Related publications (until January 2020)
- L. Cheng, M. Welborn, A. S. Christensen, T. F. Miller III, A universal density matrix functional from molecular orbital-based machine learning: Transferability across organic molecules, The Journal of Chemical Physics 150, 131103 (2019). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2 - A. S. Christensen, L. A. Bratholm, F. A. Faber, O. A. von Lilienfeld, FCHL revisited: faster and more accurate quantum machine learning, arXiv:1909.01946 (2019). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2 - A. S. Christensen, O. A. von Lilienfeld, Operator Quantum Machine Learning: Navigating the Chemical Space of Response Properties, CHIMIA 73, 1028 (2019). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2 - A. S. Christensen, F. A. Faber, O. A. von Lilienfeld, Operators in quantum machine learning: Response properties in chemical space, The Journal of Chemical Physics 150, 064105 (2019). [Open Access URL]
Dataset.
Group(s): von Lilienfeld / Project(s): INC2 - F. A. Faber, O. A. von Lilienfeld, Modeling Materials Quantum Properties with Machine Learning, 171–179 (2019).
Group(s): von Lilienfeld / Project(s): INC2 - S. Fias, K. Y. S. Chang, O. A. von Lilienfeld, Alchemical Normal Modes Unify Chemical Space, The Journal of Physical Chemistry Letters 10, 30 (2019). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2 - S. Heinen, M. Schwilk, G. F. von Rudorff, O. A. von Lilienfeld, Machine learning the computational cost of quantum chemistry, arXiv:1908.06714 (2019). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2 - S. M. Keller, M. Samarin, M. Wieser, V. Roth, Deep Archetypal Analysis, 11824, 171 (2019). [Open Access URL]
Group(s): Roth / Project(s): INC2 - A. Kortylewski, A. Wieczorek, M. Wieser, C. Blumer, S. Parbhoo, A. Morel-Forster, V. Roth, T. Vetter, Greedy Structure Learning of Hierarchical Compositional Models, 11604 (2019). [Open Access URL]
Group(s): Roth / Project(s): INC2 - P. D. Mezei, O. A. von Lilienfeld, Non-covalent quantum machine learning corrections to density functionals, arXiv:1903.09010 (2019). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2 - O. A. von Lilienfeld, K. Müller, A. Tkatchenko, Exploring Chemical Compound Space with Quantum-Based Machine Learning, arXiv:1911.10084 (2019). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2 - G. F. von Rudorff, O. A. von Lilienfeld, Rapid and accurate molecular deprotonation energies from quantum alchemy, arXiv:1911.13080 (2019). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2 - G. F. von Rudorff, O. A. von Lilienfeld, Atoms in Molecules from Alchemical Perturbation Density Functional Theory, The Journal of Physical Chemistry B 123, 10073–10082 (2019). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2 - J. Westermayr, F. A. Faber, A. S. Christensen, O. A. von Lilienfeld, P. Marquetand, Neural networks and kernel ridge regression for excited states dynamics of CH2NH2+: From single-state to multi-state representations and multi-property machine learning models, arXiv:1912.08484 (2019). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2 - P. Zaspel, B. Huang, H. Harbrecht, O. A. von Lilienfeld, Boosting Quantum Machine Learning Models with a Multilevel Combination Technique: Pople Diagrams Revisited, Journal of Chemical Theory and Computation 15, 1546–1559 (2019). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2 - T. Bereau, R. A. D. Jr., A. Tkatchenko, O. A. von Lilienfeld, Non-covalent interactions across organic and biological subsets of chemical space: Physics-based potentials parametrized from machine learning, The Journal of Chemical Physics 148, 241706 (2018). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2 - K. Y. S. Chang, O. A. von Lilienfeld, AlxGa1-xAs crystals with direct 2 eV band gaps from computational alchemy, Physical Review Materials 2, 073802 (2018). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2 - F. A. Faber, A. S. Christensen, B. Huang, O. A. von Lilienfeld, Alchemical and structural distribution based representation for universal quantum machine learning, The Journal of Chemical Physics 148, 241717 (2018). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2 - B. Huang, N. O. Symonds, O. A. von Lilienfeld, Quantum Machine Learning in Chemistry and Materials, 1–27 (2018). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2 - A. Kortylewski, M. Wieser, A. Morel-Forster, A. Wieczore, S. Parbhoo, V. Roth, T. Vetter, Informed MCMC with Bayesian Neural Networks for Facial Image Analysis, (2018). [Open Access URL]
Group(s): Roth / Project(s): INC2 - J. J. Kranz, M. Kubillus, R. Ramakrishnan, O. A. von Lilienfeld, M. Elstner, Generalized Density-Functional Tight-Binding Repulsive Potentials from Unsupervised Machine Learning, Journal of Chemical Theory and Computation 14, 2341–2352 (2018).
Group(s): von Lilienfeld / Project(s): INC2 - B. Meyer, B. Sawatlon, S. Heinen, O. A. von Lilienfeld, C. Corminboeuf, Machine learning meets volcano plots: computational discovery of cross-coupling catalysts, Chemical Science 9, 7069 (2018). [Open Access URL]
Dataset on Materials Cloud.
Group(s): Corminboeuf, von Lilienfeld / Project(s): DD1, INC2 - S. Parbhoo, M. Wieser, V. Roth, Cause-Effect Deep Information Bottleneck for Incomplete Covariates, (2018). [Open Access URL]
Group(s): Roth / Project(s): INC2 - S. Parbhoo, M. Wieser, V. Roth, Estimating Causal Effects With Partial Covariates For Clinical Interpretability, (2018). [Open Access URL]
Group(s): Roth / Project(s): INC2 - M. Rupp, O. A. von Lilienfeld, K. Burke, Guest Editorial: Special Topic on Data-Enabled Theoretical Chemistry, The Journal of Chemical Physics 148, 241401 (2018). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2 - D. N. Tahchieva, D. Bakowies, R. Ramakrishnan, O. A. von Lilienfeld, Torsional Potentials of Glyoxal, Oxalyl Halides, and Their Thiocarbonyl Derivatives: Challenges for Popular Density Functional Approximations, Journal of Chemical Theory and Computation 14, 4806–4817 (2018). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2 - O. A. von Lilienfeld, Quantum Machine Learning in Chemical Compound Space, Angewandte Chemie International Edition 57, 4164–4169 (2018).
Group(s): von Lilienfeld / Project(s): INC2 - G. F. von Rudorff, O. A. von Lilienfeld, Alchemical perturbation density functional theory for rapid yet accurate quantum property estimates, arXiv:1809.01647 (2018). [Open Access URL]
Group(s): von Lilienfeld / Project(s): INC2