A Winograd-Based Integrated Photonics Accelerator for Convolutional Neural Networks

Abstract

Neural Networks (NNs) have become the mainstream technology in the artificial intelligence (AI) renaissance over the past decade. Among different types of neural networks, convolutional neural networks (CNNs) have been widely adopted as they have achieved leading results in many fields such as computer vision and speech recognition. This success in part is due to the widespread availability of capable underlying hardware platforms. Applications have always been a driving factor for design of such hardware architectures. Hardware specialization can expose us to a novel architectural solutions, which can outperform general purpose computers for tasks at hand. Although different applications demand for different performance measures, they all share speed and energy efficiency as high priorities. Meanwhile, photonics processing has seen a resurgence due to its inherited high speed and low power nature.Here, we investigate the potential of using photonics in CNNs by proposing a CNN accelerator design based on Winograd filtering algorithm. Our evaluation results show that while a photonic accelerator can compete with current-state-of-the-art electronic platforms in terms of both speed and power, it has the potential to improve the energy efficiency by up to three orders of magnitude.

Integrated photonic FFT for photonic tensor operations towards efficient and high-speed neural networks

“…This enables de-coupling the SNR at each layer from the final output result by, for instance, electronic signal restoration, with a possible drawback introduced being latency. Nonetheless, noise is not only a limiting factor, but bears also an opportunity for NN usages; for instance, training [57] an NN with noise and performing inference tasks while dropping the absolute accuracy by about 2-3% also makes the system more robust against physical noise since the NN was conditioned with noise 'stress' [34]. Incidentally, the small-kernel algorithms such as the Winograd transformation offer an interesting alternative to the FFT filtering approach, given that many CNNs are optimized for small kernel (<13 × 13) sizes.…”

“…Given that the above mentioned high-relevance of CNNs performing the convolution operations in the optical domain is of significant interest, next we turn to scaling laws of convolution processing. Since a convolution is a high dimensional ∼N 3 problem, parallelization strategies such as multiplexing is key, which hence is synergistic to optics and photonics [32][33][34]. An interesting inspiration can be borrowed from Fourier optics [9]; instead of performing the cumbersome (allto-all) convolution between the data and the kernel, a simpler dot-product multiplication can be performed in the in the Fourier domain instead.…”

All-Chalcogenide Programmable All-Optical Deep Neural Networks

“…1a). A multi-level (2-layer) perceptron feed-forward neural network based on these novel all-optical photonic neurons, is trained using a photonic hardware model 24…”

“…1a). A multi-level (2-layer) perceptron feed-forward neural network based on these novel all-optical photonic neurons, is trained using a photonic hardware model 24 to accurately simulate the complete process and learn the weights including analog noise. The network is ultimately emulated on an open-source machine learning framework.…”

Mie Sensing with Neural Networks: Recognition of Nano‐Object Parameters, the Invisibility Point, and Restricted Models

“…During this time, advances in computer hardware have enabled the implementation of ideas developed in the field of artificial intelligence during the last century, leading to impressive demonstrations of the potential of artificial neural networks (ANNs) in machine learning (ML). These techniques have been adopted to address different problems into a variety of fields in the natural sciences, including for solving inverse problems in nanophotonics, [15,17,18] nanospectroscopy, [19] material sciences, [20,21] or microscopy. [22,23] Typically, supervised learning employing ANNs is used in two main procedures aimed at solving the inverse problem: classification [19,24] and regression.…”