In data-driven materials design where the target materials have limited data, the transfer machine learning from large known source materials, becomes a demanding strategy especially across different crystal structures. In this work, we proposed a deep transfer learning approach to predict thermodynamically stable perovskite oxides based on a large computational dataset of spinel oxides. The deep neural network (DNN) source domain model with “Center-Environment” (CE) features was first developed using the formation energy of 5329 spinel oxide structures and then was fine-tuned by learning a small dataset of 855 perovskite oxide structures, leading to a transfer learning model with good transferability in the target domain of perovskite oxides. Based on the transferred model, we further predicted the formation energy of potential 5329 perovskite structures with combination of 73 elements. Combining the criteria of formation energy and structure factors including tolerance factor (0.7 < t ≤ 1.1) and octahedron factor (0.45 < μ < 0.7), we predicted 1314 thermodynamically stable perovskite oxides, among which 144 oxides were reported to be synthesized experimentally, 10 oxides were predicted computationally by other literatures, 301 oxides were recorded in the Materials Project database, and 859 oxides have been first reported. Combing with the structure-informed features the transfer machine learning approach in this work takes the advantage of existing data to predict new structures at a lower cost, providing an effective acceleration strategy for the expensive high-throughput computational screening in materials design. The predicted stable novel perovskite oxides serve as a rich platform for exploring potential renewable energy and electronic materials applications.

We introduce an end-to-end computational framework &#xD;that allows for hyperparameter optimization &#xD;using the DeepHyper library, &#xD;accelerated model training, and interpretable &#xD;AI inference. The framework is based on state-of-the-art AI models&#xD;including CGCNN, &#xD;PhysNet, SchNet, MPNN, &#xD;MPNN-transformer, and TorchMD-NET. &#xD;We employ these AI models along with the benchmark QM9, hMOF, &#xD;and MD17 datasets to showcase how the models can predict user-specified material properties within modern &#xD;computing environments. We demonstrate &#xD;transferable applications in the modeling &#xD;of small molecules, inorganic crystals and nanoporous metal organic &#xD;frameworks with a unified, standalone framework. &#xD;We have deployed and tested this framework&#xD;in the ThetaGPU supercomputer &#xD;at the Argonne Leadership Computing Facility, &#xD;and in the Delta supercomputer at the National Center for &#xD;Supercomputing Applications to &#xD;provide researchers with modern tools to &#xD;conduct accelerated AI-driven &#xD;discovery in leadership-class computing environments. &#xD;We release these digital assets as &#xD;open source scientific software in GitLab, and &#xD;ready-to-use Jupyter notebooks in Google Colab.