2021 Fiscal Year Annual Research Report
Inverse materials design by integrating transfer learning techniques into a Bayesian framework
Project/Area Number |
18K18017
|
Research Institution | The Institute of Statistical Mathematics |
Principal Investigator |
Wu Stephen 統計数理研究所, データ科学研究系, 准教授 (70804186)
|
Project Period (FY) |
2018-04-01 – 2022-03-31
|
Keywords | polymer informatics / transfer learning / open source software |
Outline of Annual Research Achievements |
In 2021, I have finalized the new synthesis target to be liquid-crystal polyimide with high thermal conductivity based on the industrial need. Virtual library of the polyimide has been refined with the expert knowledge from our collaborators at the Tokyo Institute of Technology. More sophisticated predictive models have been trained for multiple properties to assist virtual screening with a multi-objective setting. One fifth of our proposed candidates have been successfully synthesized and under further investigations of their properties. I am preparing the final manuscript for the study.
|
Research Products
(3 results)