This serial section provides in-depth technical explanations of how machine learning algorithms solve complex physical and chemical challenges in materials science. We cover the application of the latest AI techniques, such as deep learning and generative AI, to materials exploration. In particular, we feature a special series on brainstorming and innovative ideas for constructing “Material Screening Automated Micro-Labs”—a miniaturized and optimized approach to Physical AI-based “Self-driving Labs.”
Materials science is inherently a field defined by data scarcity. Unlike image recognition or natural language processing, where tens of thousands of data points can be readily acquired, high-purity experimental…
Historically, the success of materials science has depended on how accurately human intuition could pierce through the physical and chemical principles hidden beyond the microscope. To uncover unknown properties, researchers…