Metasurfaces, which are the two-dimensional counterparts of metamaterials, have demonstrated unprecedented capabilities to manipulate the wavefront of electromagnetic waves in a single flat device. Despite various advances in this field, the unique functionalities achieved by metasurfaces have come at the cost of the structural complexity, resulting in a time-consuming parameter sweep for the conventional metasurface design. Although artificial neural networks provide a flexible platform for significantly improving the design process, the current metasurface designs are restricted to generating qualitative field distributions. In this study, we demonstrate that by combining a tandem neural network and an iterative algorithm, the previous restriction of the design of metasurfaces can be overcome with quantitative field distributions. As proof-of-principle examples, metalenses predicted via the designed network architecture that possess multiple focal points with identical/orthogonal polarisation states, as well as accurate intensity ratios (quantitative field distributions), were numerically calculated and experimentally demonstrated. The unique and robust approach for the metasurface design will enable the acceleration of the development of devices with high-accuracy functionalities, which can be applied in imaging, detecting, and sensing.

Perfect vortex beams (PVBs) possessing orbital angular momentum (OAM) and constant intensity profile enable practical applications in information encoding and transmission due to an unbounded number of orthogonal OAM channels and fixed annular intensity distributions. Geometric metasurfaces, which are 2D counterparts of metamaterials, have provided an ultra‐compact platform to flexibly design perfect vortex beams in a single flat device. However, the previous reported PVBs based on geometric metasurfaces are limited to ring‐shaped intensity profiles and intrinsic spin‐coupling between two orthogonal spin‐components. Here, spin‐decoupled geometric metasurfaces encoding with two‐step coordinate transformations are proposed to generate helicity‐independent PVBs with transmutable intensity profiles. By tailoring local phase gradient along the azimuthal direction, spin‐independent and polarization‐rotated terahertz (THz) PVBs with CN‐fold rotationally symmetric intensity profiles have been theoretically designed and experimentally demonstrated. Furthermore, THz PVBs with arbitrary intensity profiles have also been realized. The unique approach for simultaneously manipulating the spiral phase, focusing phase, as well as intensity profiles will open a new avenue to develop multifunctional integrated devices and systems, which enables potential applications in information processing and optical communication.

Benefiting from the superior capability in manipulating wavefront of electromagnetic waves, metasurfaces have provided a flexible platform for designing ultracompact and high-performance devices with unusual functionalities. As a typical functional device, multi-foci metalens can realize novel functions (i.e., the large field of view and fully reconfigurable imaging) that are extremely challenging or impossible to achieve with conventional lenses. However, a multi-foci metalens always shows inhomogeneous/chaotical intensity distributions between the multiple focal spots, which is a key challenge in metasurface design and limited to further applications. Here an iterative algorithm is proposed to automatically optimize the in-plane orientation (other than the shape) of each meta-atom in a multi-foci metalens that can generate a plethora of focal spots with uniform intensity distributions. As proof-of-principle examples, inversely designed metalenses for generating circularly-polarized, linearly-polarized, and multi-polarized images with homogeneous intensity distributions are proposed and experimentally demonstrated. The robust approach for simultaneously and accurately modulating the amplitude, phase, polarization as well as intensity distributions of terahertz waves to generate polarization-dependent and uniform intensity of focal spots will open a new avenue in developing compact imaging, face unlock, and motion sensing.