Advanced Science · 2020 · 323 citations · 29 references
ArchitectureMaterials EngineeringMaterials ScienceComputational Materials ScienceDesign OptimizationMachine LearningDeep LearningInverse DesignOptimization For Machine LearningOptimization-based DesignActive Learning DesignData-driven Materials DiscoveryGenerative Models For MaterialsInverse Material DesignInverse Materials DesignObjective Function Formulation
Machine learning is increasingly viewed as a promising tool for discovering and designing novel materials, yet the absence of robust inverse design methods that avoid exhaustive exploration remains a major bottleneck. The authors propose a general‑purpose inverse design framework that employs generative inverse design networks. The framework uses backpropagation to compute analytical gradients of an objective function with respect to design variables, enabling rapid gradient calculations and millions of optimizations from diverse initial points, and incorporates an active learning strategy to improve candidate material performance while reducing required training data, with comparisons to conventional topology optimization and genetic algorithms. The method successfully escapes local minima, generates superior designs, and reduces training data by at least an order of magnitude in a composite materials case study, outperforming passive learning and offering advantages over traditional gradient‑based and genetic optimization approaches.
Abstract In recent years, machine learning (ML) techniques are seen to be promising tools to discover and design novel materials. However, the lack of robust inverse design approaches to identify promising candidate materials without exploring the entire design space causes a fundamental bottleneck. A general‐purpose inverse design approach is presented using generative inverse design networks. This ML‐based inverse design approach uses backpropagation to calculate the analytical gradients of an objective function with respect to design variables. This inverse design approach is capable of overcoming local minima traps by using backpropagation to provide rapid calculations of gradient information and running millions of optimizations with different initial values. Furthermore, an active learning strategy is adopted in the inverse design approach to improve the performance of candidate materials and reduce the amount of training data needed to do so. Compared to passive learning, the active learning strategy is capable of generating better designs and reducing the amount of training data by at least an order‐of‐magnitude in the case study on composite materials. The inverse design approach is compared with conventional gradient‐based topology optimization and gradient‐free genetic algorithms and the pros and cons of each method are discussed when applied to materials discovery and design problems.
29