The field of additive manufacturing is moving towards the development of more accurate and efficient methods for microstructure modeling and optimization. Physics-informed machine learning approaches are emerging as a promising paradigm, allowing for the integration of physical laws into neural network architectures and enhancing accuracy, transparency, and extrapolation capabilities. These approaches have the potential to overcome the limitations of traditional experimental and computational methods, enabling predictive and scalable modeling of microstructure and its evolution across spatial and temporal scales. Noteworthy papers in this area include:

• A study on data-efficient inverse design of spinodoid metamaterials, which demonstrates the use of a neural network-based surrogate model for structure-property linkages, allowing for multi-objective inverse design with significantly less training data.
• A neural network-based computational framework for the simultaneous optimization of structural topology, manufacturable layers, and path orientations for fiber-reinforced composites, which achieves strong anisotropic strength while ensuring manufacturability.