Natural and Artificial Design of Porous Materials
|
김지한 KAIST In this talk, I will explore the trend of designing novel porous materials using artificial neural network. Specifically, I will talk about using our in-house developed generative adversarial network (GAN) software to artificially create porous materials for the first time. Moreover, we have successfully implemented inverse-design in our GAN prompting ways to train our AI to create porous materials with user-desired capacity. Next, we extend from this paradigm to integrate machine learning with genetic algorithm to screen over trillions of metal-organic frameworks (MOFs) to find record-breaking materials for methane storage. |