The use of machine learning in chemistry is on the rise for the acclerated prediction of many physicochemical properties. The input feature representation in these applications is an important factor that affects the extent of the explored chemical space. Here, we present a new descriptor, Periodic Table Tensor (PTT) descriptor that combines features from Behler-Parrinello’s symmetry functions and a periodic table representation. Using our descriptor and a Convolutional Neural Network (CNN) model, we achieved 1.9 kcal/mol and 0.07 eV/atom Mean Absolute Error (MAE) for the prediction of the total energy of organic molecules in the QM9 dataset and the formation energy of materials from Materials Project dataset, respectively. The PTT proposed in our approach enables the representation of system of any size and uses CNN to predict the energy.

Application of symmetry functions to molecule and materials using convolutional neural network

Macromolecular Chemistry: The Second Century

[Author Benefits](https://scimeetings.acs.org/?utm_source=pubs_content_marketing&utm_medium=website&utm_campaign=0320_MCF_NPI_Launch_Spring_Homepage&ref=pubs_content_marketing)

[How to use SciMeetings](https://storage.googleapis.com/dl.morressier.com/acs/SciMeetings-ACSSpring2020-HowToGuide_M3020-Final.pdf)

[Frequently Asked Questions](https://storage.googleapis.com/dl.morressier.com/acs/SciMtng_ACSNMSp2020_FAQ_M3120_Final.pdf)

ACS Spring 2020 National Meeting & Expo

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.