Materials science stories

Aug 3, 2026

Reinforcement Learning Method Advances AI-Driven Crystal Design for Functional Materials

Researchers have developed a reinforcement learning-based approach that steers generative machine learning to design novel functional materials, addressing limitations in existing crystal discovery methods that struggled to explore candidates that are both novel and useful.

Aug 3, 20261 source

Jun 30, 2026

Machine Learning Confirms Two New Superconductors, Accelerating Search for Room-Temperature Materials

International researchers used machine learning combined with quantum physics to discover and confirm two new superconducting materials—YRu₃B₂ and LuRu₃B₂—that superconduct below 1 Kelvin. The AI-driven technique dramatically accelerates the search process for superconductors, potentially bringing scientists closer to the long-sought goal of room-temperature superconductivity.

Jun 30, 20268 sources