Abstract
Developing an accurate interatomic potential model is a prerequisite for achieving reliable results from classical molecular dynamics (CMD) simulations; however, most of the potentials are biased as specific simulation purposes or conditions are considered in the parameterization. For developing an unbiased potential, a finite鈥恡emperature dynamics machine learning (FTD鈥怣L) approach is proposed, and its processes and feasibility are demonstrated using the Buckingham potential model and aluminum (Al) as an example. Compared with conventional machine learning approaches, FTD鈥怣L exhibits three distinguished features: 1) FTD鈥怣L intrinsically incorporates more extensive configurational and conditional space for enhancing the transferability of developed potentials; 2) FTD鈥怣L employs various properties calculated directly from CMD, for ML model training and prediction validation against experimental data instead of first鈥恜rinciples data; 3) FTD鈥怣L is much more computationally cost effective than first鈥恜rinciples simulations, especially when the system size increases over 103 atoms as employed in this research for ensuring reliable training data. The Al Buckingham potential developed by the FTD鈥怣L approach exhibits good performance for general simulation purposes. Thus, the FTD鈥怣L approach is expected to contribute to a fast development of interatomic potential model suitable for various simulation purposes and conditions, without limitation of model type, while maintaining experimental鈥恖evel accuracy.