Recent application of machine learning and deep learning has enabled accelerated design of artificial electromagnetic materials (AEMs), including metamaterials, metasurfaces plasmonics, frequency selective surfaces, and photonic crystals, thus overcoming significant challenges with conventional numerical methods. For many AEM systems, conventional optimization approaches are limited and not able to efficiently solve for a desired scattering response. The issue of finding the geometry of an AEM that will give a desired spectral response --- the inverse problem --- is also is of keen interest and, as of yet, still a challenging unsolved problem. Research on machine learning, deep learning for accelerated metamaterial discovery and optimization, inverse neural networks, physics informed networks, and interpretable machine learning, will be covered. A list of covered topics includes: Machine learning and deep learning for artificial electromagnetic materials; Forward and inverse deep learning approaches; Supervised learning, unsupervised learning, and reinforcement learning; Interpretable machine learning; Physics informed networks; Artificial electromagnetic materials discovery and optimization with deep learning.