Machine-Guided Understanding of Functional Oxides in Extremes

Directing the evolution of functional oxides in extreme environments is a longstanding challenge that requires new approaches to precision synthesis, characterization, and analytics. Our current inability to acquire, interpret, and act on multi-modal signatures greatly limits our control of materials for emerging applications, including quantum computing and energy storage. There is presently a transformative opportunity to harness artificial intelligence (AI) and domain-specific machine reasoning to guide the evolution of functional oxides more richly than ever before. Here I will describe our new framework for AI-guided experimentation, analytics, and modeling to explore oxides in extremes. This framework is transforming the study of fast-evolving phenomena, allowing us to discover latent defect signatures and determine key design parameters to control materials performance.