Recent breakthroughs in deep learning have ushered in an essential tool for optics and photonics, recurring in various applications of material design, system optimization, and automation control. Deep learning-enabled on-demand metasurface design and self-driving meta-devices have been the subject of extensive expansion, as it can alleviate the time-consuming and low-efficiency shortcomings in conventional numerical simulations and dependent working manner. This session focuses on the advances in how to expedite the forward prediction, inverse design, and spectral correlation of metamaterials, develop automatic meta-devices, and capture hidden physical knowledge that cannot be readily accessed. We also welcome interdisciplinary research in optics and photonics.