Multiscale modelling involves inferring physics at a given spatial and temporal scale based on the physics at a finer/smaller scale. This done under the assumption that the finer scale physics are better understood than the coarser scale physics. In this work we developed a novel neural network model called Nested Autoencoders (NestedAE) to extract important physical features and predict properties at a given length scale and correlate them with properties predicted on a larger length scale.
While this idea is general and can be applied to any system that displays distinct characteristics and properties at different length scales, we demonstrated the application of this model on
(1) a synthetic dataset created from nested analytical functions whose dimensionality is therefore known a priori, and
(2) a multi-scale metal halide perovskite dataset that is the combination of two open source datasets containing atomic and ionic properties, and device characterization using JV analysis, respectively.
For details of the datasets and how we trained NestedAE please read the paper !!paper link!!
For details on how to replicate the results in the paper please follow the instructions given in ‘replicate_results.md’’ in the docs folder.
For details on how to use NestedAE please read the user_guide.md in the docs folder.
Any questions or comments please reach out via email to the authors of the paper.