Abstract
Simultaneously accurate and efficient prediction of molecular properties throughout chemical compound space is a critical ingredient toward rational compound design in chemical and pharmaceutical industries. Aiming toward this goal, we develop and apply a systematic hierarchy of efficient empirical methods to estimate atomization and total energies of molecules. These methods range from a simple sum over atoms, to addition of bond energies, to pairwise interatomic force fields, reaching to the more sophisticated machine learning approaches that are capable of describing collective interactions between many atoms or bonds. In the case of equilibrium molecular geometries, even simple pairwise force fields demonstrate prediction accuracy comparable to benchmark energies calculated using density functional theory with hybrid exchange-correlation functionals; however, accounting for the collective many-body interactions proves to be essential for approaching the "holy grail" of chemical accuracy of 1 kcal/mol for both equilibrium and out-of-equilibrium geometries. This remarkable accuracy is achieved by a vectorized representation of molecules (so-called Bag of Bonds model) that exhibits strong nonlocality in chemical space. In addition, the same representation allows us to predict accurate electronic properties of molecules, such as their polarizability and molecular frontier orbital energies.
| Original language | English |
|---|---|
| Pages (from-to) | 2326-2331 |
| Number of pages | 6 |
| Journal | Journal of Physical Chemistry Letters |
| Volume | 6 |
| Issue number | 12 |
| DOIs | |
| Publication status | Published - 2015 Jun 18 |
Fingerprint
Keywords
- atomization energies
- chemical compound space
- machine learning
- many-body potentials
- molecular properties
ASJC Scopus subject areas
- Materials Science(all)
Cite this
Machine learning predictions of molecular properties : Accurate many-body potentials and nonlocality in chemical space. / Hansen, Katja; Biegler, Franziska; Ramakrishnan, Raghunathan; Pronobis, Wiktor; Von Lilienfeld, O. Anatole; Muller, Klaus; Tkatchenko, Alexandre.
In: Journal of Physical Chemistry Letters, Vol. 6, No. 12, 18.06.2015, p. 2326-2331.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Machine learning predictions of molecular properties
T2 - Accurate many-body potentials and nonlocality in chemical space
AU - Hansen, Katja
AU - Biegler, Franziska
AU - Ramakrishnan, Raghunathan
AU - Pronobis, Wiktor
AU - Von Lilienfeld, O. Anatole
AU - Muller, Klaus
AU - Tkatchenko, Alexandre
PY - 2015/6/18
Y1 - 2015/6/18
N2 - Simultaneously accurate and efficient prediction of molecular properties throughout chemical compound space is a critical ingredient toward rational compound design in chemical and pharmaceutical industries. Aiming toward this goal, we develop and apply a systematic hierarchy of efficient empirical methods to estimate atomization and total energies of molecules. These methods range from a simple sum over atoms, to addition of bond energies, to pairwise interatomic force fields, reaching to the more sophisticated machine learning approaches that are capable of describing collective interactions between many atoms or bonds. In the case of equilibrium molecular geometries, even simple pairwise force fields demonstrate prediction accuracy comparable to benchmark energies calculated using density functional theory with hybrid exchange-correlation functionals; however, accounting for the collective many-body interactions proves to be essential for approaching the "holy grail" of chemical accuracy of 1 kcal/mol for both equilibrium and out-of-equilibrium geometries. This remarkable accuracy is achieved by a vectorized representation of molecules (so-called Bag of Bonds model) that exhibits strong nonlocality in chemical space. In addition, the same representation allows us to predict accurate electronic properties of molecules, such as their polarizability and molecular frontier orbital energies.
AB - Simultaneously accurate and efficient prediction of molecular properties throughout chemical compound space is a critical ingredient toward rational compound design in chemical and pharmaceutical industries. Aiming toward this goal, we develop and apply a systematic hierarchy of efficient empirical methods to estimate atomization and total energies of molecules. These methods range from a simple sum over atoms, to addition of bond energies, to pairwise interatomic force fields, reaching to the more sophisticated machine learning approaches that are capable of describing collective interactions between many atoms or bonds. In the case of equilibrium molecular geometries, even simple pairwise force fields demonstrate prediction accuracy comparable to benchmark energies calculated using density functional theory with hybrid exchange-correlation functionals; however, accounting for the collective many-body interactions proves to be essential for approaching the "holy grail" of chemical accuracy of 1 kcal/mol for both equilibrium and out-of-equilibrium geometries. This remarkable accuracy is achieved by a vectorized representation of molecules (so-called Bag of Bonds model) that exhibits strong nonlocality in chemical space. In addition, the same representation allows us to predict accurate electronic properties of molecules, such as their polarizability and molecular frontier orbital energies.
KW - atomization energies
KW - chemical compound space
KW - machine learning
KW - many-body potentials
KW - molecular properties
UR - http://www.scopus.com/inward/record.url?scp=84935014439&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84935014439&partnerID=8YFLogxK
U2 - 10.1021/acs.jpclett.5b00831
DO - 10.1021/acs.jpclett.5b00831
M3 - Article
C2 - 26113956
AN - SCOPUS:84935014439
VL - 6
SP - 2326
EP - 2331
JO - Journal of Physical Chemistry Letters
JF - Journal of Physical Chemistry Letters
SN - 1948-7185
IS - 12
ER -