Polymer Design via Data-Driven Approaches
Data-driven approaches, trained with massive amount of data, has gain huge success in various fields, e.g., material design. Polymer, as a popular material that is ubiquitous in different kinds of applications, has been intensively explored and designed based on experiments and simulations in the past. As more and more polymer data are accumulated across the history, data-driven approaches, known for its capability in fast pattern recognition, have been employed to accelerate the design of polymers. In this presentation, polymer design via data-driven approaches will be systematically investigated, including polymer representation, quantitative structure-property relationship, high-throughput molecular dynamics simulation for polymer labeling and property validation, inverse polymer design, and synthetic accessibility estimation.
Dr. Ruimin Ma, obtains his Ph.D. in 2021 from the University of Notre Dame, United States, with a research focus on polymer informatics. His research interests spread out broadly to many other AI related topics, like transfer learning, reinforcement learning, multi-agent learning, generative modeling, etc. Before his Ph.D., he was an undergraduate at Huazhong University of Science and Technology, with a research focus on molecular dynamics simulations.