Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userScalability of all-optical neural networks based on spatial light modulators
193
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
Optical implementation of artificial neural networks has been attracting great attention due to its potential in parallel computation at speed of light. Although all-optical deep neural networks (AODNNs) with a few neurons have been experimentally demonstrated with acceptable errors recently, the feasibility of large scale AODNNs remains unknown because error might accumulate inevitably with increasing number of neurons and connections. Here, we demonstrate a scalable AODNN with programmable linear operations and tunable nonlinear activation functions. We verify its scalability by measuring and analyzing errors propagating from a single neuron to the entire network. The feasibility of AODNNs is further confirmed by recognizing handwritten digits and fashions respectively.
rate research
Read More
Deeplearning algorithms are revolutionising many aspects of modern life. Typically, they are implemented in CMOS-based hardware with severely limited memory access times and inefficient data-routing. All-optical neural networks without any electro-optic
Software-implementation, via neural networks, of brain-inspired computing approaches underlie many important modern-day computational tasks, from image processing to speech recognition, artificial intelligence and deep learning applications. Yet, differing from real neural tissue, traditional computing architectures physically separate the core computing functions of memory and processing, making fast, efficient and low-energy brain-like computing difficult to achieve. To overcome such limitations, an attractive and alternative goal is to design direct hardware mimics of brain neurons and synapses which, when connected in appropriate networks (or neuromorphic systems), process information in a way more fundamentally analogous to that of real brains. Here we present an all-optical approach to achieving such a goal. Specifically, we demonstrate an all-optical spiking neuron device and connect it, via an integrated photonics network, to photonic synapses to deliver a small-scale all-optical neurosynaptic system capable of supervised and unsupervised learning. Moreover, we exploit wavelength division multiplexing techniques to implement a scalable circuit architecture for photonic neural networks, successfully demonstrating pattern recognition directly in the optical domain using a photonic system comprising 140 elements. Such optical implementations of neurosynaptic networks promise access to the high speed and bandwidth inherent to optical systems, which would be very attractive for the direct processing of telecommunication and visual data in the optical domain.
Liquid crystal based spatial light modulators are widely used in applied optics due to their ability to continuously modulate the phase of a light field with very high spatial resolution. A common problem in these devices is the pixel crosstalk, also called the fringing field effect, which causes the response of these devices to deviate from the ideal behavior. This fringing effect decreases the performance of the spatial light modulator and is shown to cause an asymmetry in the diffraction efficiency between positive and negative diffraction orders. We use simulations of the director distribution to reproduce diffraction efficiency measurements of binary and blazed gratings. To overcome these limitations in performance, the simulations of the director distribution in the liquid crystal layer are used to develop a fast and precise model to compute the phase response of the spatial light modulator. To compensate the fringing field effect, we implement this model in phase retrieval algorithms and calculate the phase profile corresponding to a regular spot pattern as a generic example. With this method, we are able to increase the spot uniformity significantly compared to a calculation without considering the fringing field effect. Additionally, polarization conversion efficiencies of various simple phase patterns are simulated and measured for different orientations of the spatial light modulator. We found that the polarization conversion has the the smallest effect for a setup in which the liquid crystal molecules at the alignment layer lie in the plane of incidence of the light beam.
Spatial light modulators (SLMs) are central to numerous applications ranging from high-speed displays to adaptive optics, structured illumination microscopy, and holography. After decades of advances, SLM arrays based on liquid crystals can now reach large pixel counts exceeding 10^6 with phase-only modulation with a pixel pitch of less than 10 {mu}m and reflectance around 75%. However, the rather slow modulation speed in such SLMs (below hundreds of Hz) presents limitations for many applications. Here we propose an SLM architecture that can achieve high pixel count with high-resolution phase-only modulation at high speed in excess of GHz. The architecture consists of a tunable two-dimensional array of vertically oriented, one-sided microcavities that are tuned through an electro-optic material such as barium titanate (BTO). We calculate that the optimized microcavity design achieves a {pi} phase shift under an applied bias voltage below 10 V, while maintaining nearly constant reflection amplitude. As two model applications, we consider high-speed 2D beam steering as well as beam forming. The outlined design methodology could also benefit future design of spatial light modulators with other specifications (for example amplitude modulators). This high-speed SLM architecture promises a wide range of new applications ranging from fully tunable metasurfaces to optical computing accelerators, high-speed interconnects, true 2D phased array beam steering, and quantum computing with cold atom arrays.
For the benefit of designing scalable, fault resistant optical neural networks (ONNs), we investigate the effects architectural designs have on the ONNs robustness to imprecise components. We train two ONNs -- one with a more tunable design (GridNet) and one with better fault tolerance (FFTNet) -- to classify handwritten digits. When simulated without any imperfections, GridNet yields a better accuracy (~98%) than FFTNet (~95%). However, under a small amount of error in their photonic components, the more fault tolerant FFTNet overtakes GridNet. We further provide thorough quantitative and qualitative analyses of ONNs sensitivity to varying levels and types of imprecisions. Our results offer guidelines for the principled design of fault-tolerant ONNs as well as a foundation for further research.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
