Fig. 1: An overview of the proposed workflow (GAD-MALL, i.e., Generative architecture design—multi-objective active learning Loop).
From: Machine learning-enabled constrained multi-objective design of architected materials
a The neural network proposes candidates with unknown properties. b The machine-learning (ML) algorithm interactively queries the finite element methods (FEM) to propose new designs. c The 3D printing technique fabricates the proposed architectural design. d GAD-MALL explores the design landscape of architected materials and discovers various high-performance architected materials (mean ± SD, n = 3). Source data are provided as a Source Data file.
