Deep Reasoning Networks: Combining deep learning with reasoning for discovery

Crystal-structure phase mapping is a core, long-standing challenge in materials science that requires identifying crystal phases, or mixtures thereof, in x-ray diffraction measurements of synthesized materials. Phase mapping algorithms have been developed that excel at solving systems with up to several unique phase mixtures, wherein each phase has a readily distinguishable diffraction pattern. However, complexities such as dozens of phase mixtures, alloy-dependent variation in diffraction patterns, and multiple compositional degrees of freedom pose challenges for materials science experts and state-of-the-art algorithms, creating a major bottleneck in high-throughput materials discovery. Herein we show how to automate crystal-structure phase mapping. We formulate phase mapping as an unsupervised pattern demixing problem and describe how to solve it using Deep Reasoning Networks (DRNets). Given the scientific complexity of crystal-structure phase mapping, we also provide an intuitive explanation of DRNets framework based on Multi-MNIST-Sudoku, a variant of the Sudoku game that involves demixing two completed overlapping hand-written Sudokus. DRNets combine deep learning with constraint reasoning for incorporating prior scientific knowledge and consequently require only a modest amount of (unlabeled) data. DRNets compensate for the limited data by exploiting and magnifying the rich prior-knowledge about the thermodynamic rules governing the mixtures of crystals. DRNets are designed with an interpretable latent space for encoding prior-knowledge domain constraints and seamlessly integrate constraint reasoning into neural network optimization. DRNets surpass previous approaches on crystal-structure phase mapping, unraveling the Bi-Cu-V oxide phase diagram, and aiding the discovery of solar-fuels materials.
For more information, see:
Di Chen, Yiwei Bai, Sebastian Ament, Wenting Zhao, Dan Guevarra, Lan Zhou, Bart Selman, R. Bruce van Dover, John M. Gregoire, Carla P. Gomes. Automating crystal-structure phase mapping by combining deep learning with constraint reasoning. Nature Machine Intelligence (2021). www.nature.com...