An Optical Frontend for a Convolutional Neural Network
This addresses energy consumption issues in AI hardware for image processing, though it is incremental as it combines existing optical and electronic methods.
The paper tackled the problem of energy inefficiency in optical convolutional neural networks by designing a hybrid photonic-electronic architecture that uses a free-space optical frontend for the first layer, achieving 87% accuracy on a modified AlexNet for the Kaggle Cats and Dogs dataset.
The parallelism of optics and the miniaturization of optical components using nanophotonic structures, such as metasurfaces present a compelling alternative to electronic implementations of convolutional neural networks. The lack of a low-power optical nonlinearity, however, requires slow and energy-inefficient conversions between the electronic and optical domains. Here, we design an architecture which utilizes a single electrical to optical conversion by designing a free-space optical frontend unit that implements the linear operations of the first layer with the subsequent layers realized electronically. Speed and power analysis of the architecture indicates that the hybrid photonic-electronic architecture outperforms sole electronic architecture for large image sizes and kernels. Benchmarking of the photonic-electronic architecture on a modified version of AlexNet achieves a classification accuracy of 87% on images from the Kaggle Cats and Dogs challenge database.