Recurrent Neural Network-based Model for Accelerated Trajectory Analysis in AIMD Simulations
This work addresses the computational bottleneck of AIMD simulations for materials researchers, but it is incremental as it applies existing RNN methods to a new domain.
The paper tackles the problem of accelerating trajectory analysis in ab initio molecular dynamics (AIMD) simulations by using recurrent neural networks (RNNs) to forecast trajectory paths and potential energy profiles from atom coordinate distributions, demonstrating that both gated recurrent unit and long short-term memory networks can potentially enable accelerated statistical sampling in computational materials research.
The presented work demonstrates the training of recurrent neural networks (RNNs) from distributions of atom coordinates in solid state structures that were obtained using ab initio molecular dynamics (AIMD) simulations. AIMD simulations on solid state structures are treated as a multi-variate time-series problem. By referring interactions between atoms over the simulation time to temporary correlations among them, RNNs find patterns in the multi-variate time-dependent data, which enable forecasting trajectory paths and potential energy profiles. Two types of RNNs, namely gated recurrent unit and long short-term memory networks, are considered. The model is described and compared against a baseline AIMD simulation on an iridium oxide slab. Findings demonstrate that both networks can potentially be harnessed for accelerated statistical sampling in computational materials research.