Creating future materials using machine learning

Creating future materials using machine learning

As the world sees rapid improvements in areas such as renewable energy, water management, transportation, and quantum computing, the need for new types of advanced materials is growing. However, developing new materials is tedious – it takes decades from concept to implementation.

For the rapid discovery of materials, the manufacturing sector is now shifting to artificial intelligence (AI) and machine learning techniques. But for those looking for a job in this bold new material engineering, there are still a few postgraduate options designed to integrate materials with machine learning.

To address this significant knowledge gap, the USC Whiteby School of Engineering Mork Family Department of Chemical Engineering and Materials Science has introduced the new MS in Material Engineering, a unique science master’s course, a course taught by experts in Materials Science and Machine Learning.

Andrea Hodge, Arthur B. Freeman, professor and chair of the Mork family of chemical engineering and materials science, said the course reflects a new multidisciplinary field that uses information and modeling and examples, experiments and machine learning technologies. Material discovery and production approach.

There are currently many online courses in material engineering and many aspects of machine learning, but there is no MS degree program that offers integration of machine learning into material engineering, he said.

“Increasingly, companies are incorporating information-based approaches into their business models,” he said. However, due to the lack of integrated machine learning, the current material engineering curriculum is not meeting the demand.

Machine learning can be used to create 3D-printing, water filtration, quantum data and computer materials and covers, and to find new materials for a variety of applications to speed up automation and processes.

Machine learning can also be used in research and development, such as automated laboratories and advanced material filters based on databases, high energy scales, novel stimuli, improved solar panels, and new drugs to fight viruses and diseases.

Ken-Aichi Numera, a senior lecturer at the Mork Family Department, said machine learning would make a symbolic difference in material engineering.

“The goal is not to strengthen existing skills with new skills to accelerate material development and innovation,” he said.

The new master’s degree program focuses on material education that will allow future material scientists and engineers to use the vast amount of information that humans cannot afford to process.

According to Numera, machine learning has made great strides in areas such as computer vision – where systems can learn information from digital images and videos – as well as natural language processing, process automation and more.

“However, in order to fully utilize the power of these new technologies, the knowledge of the domain expert is necessary because machine learning technologies do not transfer directly to materials,” he said. “The new Master’s Degree Program will train future generations of manpower and provide exciting opportunities for USC graduates in industries that include information-based discovery and material development.

Beginning in late 2021, the program will focus on graduates with a broad science and engineering background at the crossroads of material and machine learning – including BS students in physical science and engineering. The program is for those in the industry who want to expand their skill sets and apply machine learning techniques to research and development efforts in materials and processing projects.

The program covers the following key areas:

* Basics of machine learning, including supervised and uncontrolled machine learning, regression and placement, and active learning and reinforcement.

* Deep learning methods and their applications.

* Simulation and machine learning projects to use databases for discovery and model material features.

At the annual Mork Family School Student Symposium, students will also have the opportunity to present results from their projects.

“There are also opportunities for machine learning as well as for start-up research in universities, industry and national laboratories,” said Professor Pry Vashta, chairman of Flow Flower in Chemical Engineering.

The Mork Family Department faculty, which teaches at the program, includes high-performance simulators and machine learning researchers, as well as material genomics focused on mathematical modeling for computer software center director Vashta. And machine learning. Rajiv Kalia, a professor of physics and astronomy, computer science, chemical engineering and material science, teaches in-depth learning techniques and applications.

Additional material science, machine learning, and data analysis courses taught by MFD and USC Vitterbi are included as part of the elective courses to provide extensive learning experience in machine learning in material engineering.

Photo or key art available *

Yes, see below.

Photo / Key Art Attachment # 1


320.40 KB? PNG

# 1 Photo / Key Art Description

A state-of-the-art neural network (RNA) model is being used to estimate a better polymer structure with higher energy storage capacity.

# 1 Photo / Key Art Acknowledgments

Chris Kim

/ Public statement. This article may be from an original source and may be subject to clarity, style, and preparation. See the whole thing here.


Leave a Comment