Combining AI Reasoning and Machine Learning for Accelerating Materials Discovery

Artificial Intelligence (AI) is a rapidly advancing field inspired by human intelligence. AI systems are now performing at human and even superhuman levels on various tasks, such as image identification and face and speech recognition. The tremendous AI progress that we have witnessed in the last decade has been primarily driven by deep learning advances and heavily hinges on the availability of large, annotated datasets to supervise model training. However, often we only have access to small datasets and incomplete data. Humans amplify a few data examples with intuitions and detailed reasoning from first principles and prior knowledge for discovery. I will describe Deep Reasoning Networks (DRNets), a general framework that seamlessly integrates deep learning and reasoning via an interpretable latent space for incorporating prior knowledge and tackling challenging problems. DRNets requires only modest amounts of (unlabeled) data, in sharp contrast to standard deep learning approaches. DRNets reach superhuman performance for crystal-structure phase mapping, a core, long-standing challenge in materials science, enabling the discovery of solar-fuels materials. For an intuitive demonstration of our approach, we also solve variants of the Sudoku game. This work was featured in a cover article of Nature Machine Intelligence entitled, Automating crystal-structure phase mapping by combining deep learning with constraint reasoning. I will also talk about the Scientific Autonomous Reasoning Agent (SARA) for accelerating materials discovery, which was recently featured in a Science Advances article entitled, Autonomous materials synthesis via hierarchical active learning of non-equilibrium phase diagrams.