Spatially incoherent diffractive optical processors can handle data beyond non-negative values, potentially making them valuable in diverse scenarios, such as visual encryption and autonomous vehicle ...

The proposed diffractive deep neural network employs orbital angular momentum encoding and diffractive layers to process spatial information from handwritten digits, offering a robust and versatile ...

They revealed that simpler-to-implement phase encoding matches the accuracy of data repetition strategies across various test datasets. While data repetition-based diffractive blocks cannot provide ...

Layered metasurfaces trained as optical neural networks enable multifunctional holograms and security features, integrating neural computation principles with nanostructured optics to create a ...

In this recurring monthly feature, we filter recent research papers appearing on the arXiv.org preprint server for compelling subjects relating to AI, machine learning and deep learning – from ...

State-of-the-art neural networks depend on linear operations, such as matrix-vector multiplications and convolutions. While dedicated processors like GPUs and TPUs exist for these operations, they ...