Meta-Learning Optimization for Accelerated Materials Discovery

The project integrates state-of-the-art meta-learning, cost-aware experimental design, and multiscale materials engineering to drastically reduce trial-and-error in laboratory research. Led by experts in AI and ceramic materials, we aim to automatically explore process–structure–property relationships and deliver new “multi-tough” ceramics—materials with significantly enhanced fracture resistance—alongside a transferable AI platform for broader materials innovation.