|
|
| Type |
Award Lecture in Division |
| Area |
Recent Trend in Surface and Interface Physical Chemistry |
| Room No. |
Room 208+209+210 |
| Time |
FRI 14:30-: |
| Code |
PHYS3-1 |
| Subject |
Accelerating Materials Discovery with Scalable Computations and Machine Learning |
| Authors |
Yousung Jung
Korea Advanced Institute of Science and Technology, Korea |
| Abstract |
Novel materials discovery is a key to addressing many challenges in energy, climate change, and future sustainability. Usual procedure of finding innovative materials based mainly on experiments, however, can take far too long due to a vast and discrete search space, and thus accelerating this process by orders of magnitude using scalable computations would significantly reduce the time and cost of new discovery. In achieving this grand goal, density functional first principles simulation offers a sweet spot between the prediction accuracy and feasibility for most current large scale materials applications. In this lecture, I will talk about some of our recent efforts along this line to understand and design new materials that capture carbon, catalyze CO2 conversion, and provide storage of that energy. All of these materials are complex, but can also be potentially highly tunable if there is a way to accurately extrapolate the large set of existing data for a new discovery, precisely the area in which the machine learning and artificial intelligence has made waves and significant breakthroughs in recent years. At the end of this talk, thus, I will briefly describe great opportunities in which state-of-the-art computer science techniques in machine learning can contribute greatly to creating solutions to materials problems.
|
| E-mail |
ysjn@kaist.ac.kr |
|
120th General Meeting of the KCS