Machine learning force fields: Construction, validation, and outlook
Venkatesh Botu, Rohit Batra, James Chapman, Rampi Ramprasad (Submitted on 6 Oct 2016)
Force fields developed with machine learning methods in tandem with quantum mechanics are beginning to find merit, given their (i) low cost, (ii) accuracy, and (iii) versatility. Recently, we proposed a force based approach, wherein, the vectorial force on an atom is computed directly from its environment. Here, we discuss a multi-step workflow for their construction, which begins with generating diverse reference atomic environment and force data, choosing a representation for the atomic environments, down selecting a representative set, and lastly the learning method itself, for the case of Al. Further, methods to judge force prediction accuracy are proposed allowing for an adaptive refinement of the force field. The constructed force field is then validated by simulating complex materials phenomena such as melting and stress-strain behavior, which truly go beyond the realm of ab initio methods both in length and time scales.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.