Shao-Chun Lee, Ph.D. Candidate
Dr. Y Z (Yang Zhang), Director of Research
March 25, 2024 | 900am - 11:00am CST
This final examination will be held in 101A Talbot Laboratory.
Investigations into liquid physics in energy applications with machine-learning interatomic potentials: An active learning framework utilizing Subascent
Abstract: Liquids are one of the fundamental states of matter whose properties are frequently investigated through atomistic scale simulations as a comprehensive theoretical framework for liquids is lacking. In the past two decades, machine-learning interatomic potentials (MLIPs) have emerged as a promising tool for modeling liquids, yet a significant remaining challenge lies in the model’s accuracy and reliability being highly contingent upon the diversity of the training data.
In this research, we proposed an active learning framework that iteratively trains a MLIP with new training data sampled via Subascent, an algorithm to efficiently sample the rare events with minimal domain-specific knowledge. To further enhance the sampling efficiency, we not only expedited the convergence rate of Subascent by one to two orders of magnitude but also enhanced the convergence accuracy. The training data for MLIPs, originally limited to only near-equilibrium data, can be augmented with rare yet crucial transition states which is beyond the reach of conventional MD simulations. Furthermore, the capability to uncover these essential collective motions without any a priori knowledge is unprecedented.