Stanford University’s Evan Reed is a PhD in Physics from MIT. His team researches on nano-scale materials in electronics and energy applications. The team is reported to be applying machine learning to “develop better electrolytes for lithium-ion batteries”. Since electrolytes constitute a combination of materials, identifying the right mix is a challenge to the human brain. In comes the technology called Machine Learning.
“We have developed a machine learning model that has been outperforming experts’ intuition when predicting which materials to use.” Prof. Reed says.
Machine Learning has seen application in Superconductor Research, Development of Thermo-electric Materials, Hydrogen Storage Units (for fuel cells), Nanotube Construction, Pharmaceuticals.
Machine Learning comes in handy to improve processes. Shyam Dwarkanath, Materials Research Scientist (Berkley Lab), has done lot of work in materials having applications in solar, energy harvesting, efficiency and more. He has been quoted as saying;
“Many processes in materials science rely on some sort of classification or fitting. Traditionally, this has been done by hand or some simple linear model after significant data processing. Machine learning makes these tasks much easier while improving the quality, speed, and amount of data that can be extracted. This has yielded automated methods for constructing phase diagrams, predicting structures for new compositions, and even analyzing micrographs in place of humans.”
The Where-to-Apply question?
Application of Machine Learning in material development is at a very early stage. Scientists are trying to figure out where specifically Machine Learning can come in handy. With data being the crux for decision making here, data of huge proportions are being assimilated to make application of Machine Learning more rewarding. Data will answer the application question sooner or later. The data will provide the researchers excellent insights on the material. The next challenge will be to come with an application of the New Material especially for the industrial use. And that’s where Machine Learning will kick in again.
The Future of Machine Learning in Material Sciences
The research community is positive about the prospects of material science given the backing of Machine Learning.
Valentin Stanev is a Research Scientist at University of Maryland. He focuses on “applying machine learning and data mining methods to material science problems”. According to him;
“You can have a machine learning toolbox built into your experimental setup. It looks at the results coming out of the experiment and can algorithmically decide what experiment to do next and from these deduce the general outcome of a series of experiments. In a way, you may only need to run 10 or 20% of the experiment to get the full picture”
Evan Reed believes that machine learning can help work backwards to arrive at a material of a certain spec. He says “Imagine that you need a battery that has a certain set of properties. You feed those into the machine learning model that then automatically runs through all available, known materials and suggests a range of batteries consisting of different materials that meet your specifications.”