Abstract:Machine learning promises to accelerate materials discovery by allowing computational efficient property predictions from a small number of reference calculations. As a result, the literature has spent a considerable effort in designing representations that capture basic physical properties. Our work focuses on the less ‐studied learning formulations in this context in order to exploit inner structures in the prediction errors. In particular, we propose to directly optimize basic loss functions of the predictio… Show more
“…SLATM is composed of two-and three-body potentials, which are derived from the atomic coordinates and contain most of the relevant information to predict molecular properties. 70,[88][89][90][91][92][93][94] (3) Training of the model: input representations are mapped onto the corresponding target values (E a , computed at the DFT level, see the next section) using Kernel Ridge Regression (KRR) 95 with a Gaussian kernel. Note that even if target values based on DFT are used here to train the ML model, the strategy proposed hereaer is expected to perform equally well on experimental or more accurate quantum chemical data.…”
HHundreds of catalytic methods are developed each year to meet the demand for high-purity chiral compounds. The computational design of enantioselective organocatalysts remains a significant challenge, as catalysts are typically...
“…SLATM is composed of two-and three-body potentials, which are derived from the atomic coordinates and contain most of the relevant information to predict molecular properties. 70,[88][89][90][91][92][93][94] (3) Training of the model: input representations are mapped onto the corresponding target values (E a , computed at the DFT level, see the next section) using Kernel Ridge Regression (KRR) 95 with a Gaussian kernel. Note that even if target values based on DFT are used here to train the ML model, the strategy proposed hereaer is expected to perform equally well on experimental or more accurate quantum chemical data.…”
HHundreds of catalytic methods are developed each year to meet the demand for high-purity chiral compounds. The computational design of enantioselective organocatalysts remains a significant challenge, as catalysts are typically...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.