• Neural Network Kinetics

Our research focuses on fundamentally understanding material behaviors under extreme environments, including high stress, elevated temperatures, and intense radiation flux. To address the timescale challenges inherent in atomistic modeling, we develop computational and simulation algorithms, such as energy landscape sampling , neural network kinstics (NKK) to reveal slow defect kinetics (e.g., vacancy diffusion correlation ). By integrating cutting-edge electron microscopy, we study how strain localization and plasticity evolve under stress- and time-dependent driven conditions in heterogeneous, non-equilibrium systems (e.g., maximum strength , slip banding ). This combined approach enables us to tune atomic-scale mechanisms and engineer advanced alloys designed for high-performance nuclear energy and aerospace applications.