Novel machine learning based framework could lead to breakthroughs in material design

October 10, 2018, Virginia Tech
Yaxin An, Sanket A. Deshmukh, and Karteek Bejagam. Credit: Virginia Tech

Computers used to take up entire rooms. Today, a two-pound laptop can slide effortlessly into a backpack. But that wouldn't have been possible without the creation of new, smaller processors—which are only possible with the innovation of new materials.

But how do scientists actually invent new materials? Through experimentation, explains Sanket Deshmukh, an assistant professor in the chemical engineering department whose team's recently published computational research might vastly improve the efficiency and costs savings of the material design process.

Deshmukh's lab, the Computational Design of Hybrid Materials lab, is devoted to understanding and simulating the ways molecules move and interact—crucial to creating a new material.

In recent years, machine learning, a powerful subset of artificial intelligence, has been employed by materials scientists to accelerate the discovery of through computer simulations. Deshmukh and his team have recently published research in the Journal of Physical Chemistry Letters demonstrating a novel machine learning that trains "on the fly," meaning it instantaneously processes data and learns from it to accelerate the development of computational models.

Traditionally the development of computational models are "carried out manually via trial-and-error approach, which is very expensive and inefficient, and is a labor-intensive task," Deshmukh explained.

"This novel framework not only uses the machine learning in a unique fashion for the first time," Deshmukh said, "but it also dramatically accelerates the development of accurate computational models of materials."

"We train the machine learning model in a 'reverse' fashion by using the properties of a model obtained from as an input for the machine learning model, and using the input parameters used in molecular dynamics simulations as an output for the machine learning model," said Karteek Bejagam, a post-doctoral researcher in Deshmukh's lab and one of the lead authors of the study.

This new framework allows researchers to perform optimization of computational models, at unusually faster speed, until they reach the desired properties of a new material.

The best part? Regardless of how accurate the predictions of machine learning models are, as they are tested on-the-fly, these models have no negative impact on the model optimization, if it's inaccurate. "It can't hurt, it can only help," said Samrendra Singh, a visiting scholar in Deshmukh's lab and another author of the study.

"The beauty of this new machine learning framework is that it is very general, meaning the machine learning can be integrated with any optimization algorithm and computational technique to accelerate the materials design," Singh said.

The publication, lead by Bejagam and Singh and with the collaboration of chemical engineering Ph.D. student Yaxin An, shows the use of this new framework by developing the models of two solvents as a proof of concept.

Deshmukh's lab plan to build on the research by utilizing this novel machine learning based framework to develop models of various materials that have potential biomedicine and energy applications.

Explore further: Machine learning enables predictive modeling of 2-D materials

More information: Karteek K. Bejagam et al. Machine-Learned Coarse-Grained Models, The Journal of Physical Chemistry Letters (2018). DOI: 10.1021/acs.jpclett.8b01416

Related Stories

Artificial intelligence may put private data at risk

August 3, 2018

Machine learning – a form of artificial intelligence in which computers use data to learn on their own – is rapidly growing and poised to transform the world. But current models are vulnerable to privacy leaks and other ...

Accelerating electrocatalyst discovery with machine learning

September 13, 2018

Researchers are paving the way to total reliance on renewable energy as they study both large- and small-scale ways to replace fossil fuels. One promising avenue is converting simple chemicals into valuable ones using renewable ...

Recommended for you

Carbon fuels go green for renewable energy

December 18, 2018

For decades, scientists have searched for effective ways to remove excess carbon dioxide emissions from the air, and recycle them into products such as renewable fuels. But the process of converting carbon dioxide into useful ...

0 comments

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.