Dr. Han Zhongkang from Max Planck Institute of Germany came to the center to give a special report

Time: 2021-6-10    Source：Collaborative Innovation Center

On June 10th, from 10:00 to 11:30, Dr. Han Zhongkang from Max Planck Institute in Germany was invited to the Co-Creation Center to conduct a lecture entitled "Accurate simulation of phase diagrams and phase transition processes of materials under real conditions and machine learning in high performance Academic lecture on "Applications in Material Design Screening".

In the lecture, Dr. Han mentioned that machine learning has been widely used in the field of materials. It has developed rapidly in the past few years and has become one of the most popular materials design methods for materials researchers. However, due to the complexity of the model machine trained by the deep learning method, it is difficult to provide new understanding and knowledge of the research content or system, and deep learning requires a large number of samples to provide training for machine learning. Dr. Han proposed a new type of machine learning software developed by their research group, using numerical learning machine learning methods to train models with sufficiently high accuracy and clearer physical mechanisms. Through this model, researchers can obtain the key factors affecting high-performance materials, so as to realize the reasonable design and selection of high-performance materials.

After the meeting, Dr. Han had a lively discussion with teachers and students, and everyone showed strong recognition and interest in machine learning methods. In addition, Professor Xu Jinwu put forward some novel ideas and very valuable suggestions for some current situation and problems in the field of machine learning. The academic lecture ended successfully, and the teachers and students burst into new sparks during the exchanges and discussions, which benefited a lot.