Machine-learning-based one-to-many inverse design of multi-material lattices

This work presents an ML-based inverse design framework for multi-material lattices with curved struts, targeting mechanical and thermal performance. Using cubic-spline parameterization and discrete material assignment, the design space expands beyond conventional lattices. A workflow combining a material classifier, property predictor, and inverse generators addresses one-to-many mapping, enabling probabilistic sampling and diverse designs. The approach supports multi-objective trade-offs and lays the foundation for multi-scale optimization of functionally graded metamaterials.