Skip to Main Content U.S. Department of Energy
Science Directorate
Page 12 of 278

Advanced Computing, Mathematics and Data

May 2018

Chemception Takes a Deeper Look at Chemical Behavior

In a recent conversation with Chemistry World about how deep learning (machine learning algorithms based on layered neural networks) is changing how researchers are exploring chemical behavior, ACMD Division Director Nathan Baker discussed Chemception, a deep convolutional neural network framework developed to predict a molecule’s chemical properties from its structure.

Unlike existing models that perform similar functions and have been built with decades of chemical knowledge, Chemception builds its own models using only two-dimensional pictures of molecules. According to Baker, such capability has potential applications in areas such as biophysics.

Read more in the Chemistry World article: “Machine learning hits the big screen.

Page 12 of 278

Science at PNNL

Core Research Areas

User Facilities

Centers & Institutes

Additional Information

Research Highlights Home


Print this page (?)

YouTube Facebook Flickr TwitThis LinkedIn