Deep learning with coherent nanophotonic circuits
01 Jul 2017-Vol. 11, Iss: 7, pp 441-446
TL;DR: A new architecture for a fully optical neural network is demonstrated that enables a computational speed enhancement of at least two orders of magnitude and three order of magnitude in power efficiency over state-of-the-art electronics.
Abstract: Artificial Neural Networks have dramatically improved performance for many machine learning tasks. We demonstrate a new architecture for a fully optical neural network that enables a computational speed enhancement of at least two orders of magnitude and three orders of magnitude in power efficiency over state-of-the-art electronics.
Citations
More filters
Posted Content•
TL;DR: In this paper, the authors proposed an acceleration method for learning MZI parameters by creating customized complex-valued derivatives for an MZI, exploiting Wirtinger derivatives and a chain rule.
Abstract: An optical neural network (ONN) is a promising system due to its high-speed and low-power operation. Its linear unit performs a multiplication of an input vector and a weight matrix in optical analog circuits. Among them, a circuit with a multiple-layered structure of programmable Mach-Zehnder interferometers (MZIs) can realize a specific class of unitary matrices with a limited number of MZIs as its weight matrix. The circuit is effective for balancing the number of programmable MZIs and ONN performance. However, it takes a lot of time to learn MZI parameters of the circuit with a conventional automatic differentiation (AD), which machine learning platforms are equipped with. To solve the time-consuming problem, we propose an acceleration method for learning MZI parameters. We create customized complex-valued derivatives for an MZI, exploiting Wirtinger derivatives and a chain rule. They are incorporated into our newly developed function module implemented in C++ to collectively calculate their values in a multi-layered structure. Our method is simple, fast, and versatile as well as compatible with the conventional AD. We demonstrate that our method works 20 times faster than the conventional AD when a pixel-by-pixel MNIST task is performed in a complex-valued recurrent neural network with an MZI-based hidden unit.
3 citations
TL;DR: In this article , the authors proposed an optical subspace neural network (OSNN) architecture, which trades the universality of weight representation for lower optical component usage, area cost, and energy consumption.
Abstract: The optical neural network (ONN) is a promising hardware platform for next-generation neurocomputing due to its high parallelism, low latency, and low energy consumption. Previous ONN architectures are mainly designed for general matrix multiplication (GEMM), leading to unnecessarily large area cost and high control complexity. Here, we move beyond classical GEMM-based ONNs and propose an optical subspace neural network (OSNN) architecture, which trades the universality of weight representation for lower optical component usage, area cost, and energy consumption. We devise a butterfly-style photonic–electronic neural chip to implement our OSNN with up to 7× fewer trainable optical components compared to GEMM-based ONNs. Additionally, a hardware-aware training framework is provided to minimize the required device programming precision, lessen the chip area, and boost the noise robustness. We experimentally demonstrate the utility of our neural chip in practical image recognition tasks, showing that a measured accuracy of 94.16% can be achieved in handwritten digit recognition tasks with 3 bit weight programming precision.
3 citations
TL;DR: In this article , the phase-encoding and intensity detection is used to solve arbitrary Ising problems on demand, which is based on the simulated annealing algorithm and requires only one step of optical linear transformation with simplified Hamiltonian calculation.
Abstract: Photonic Ising machine is a new paradigm of optical computing, which is based on the characteristics of light wave propagation, parallel processing and low loss transmission. Thus, the process of solving the combinatorial optimization problems can be accelerated through photonic/optoelectronic devices. In this work, we have proposed and demonstrated the so-called Phase-Encoding and Intensity Detection Ising Annealer (PEIDIA) to solve arbitrary Ising problems on demand. The PEIDIA is based on the simulated annealing algorithm and requires only one step of optical linear transformation with simplified Hamiltonian calculation. With PEIDIA, the Ising spins are encoded on the phase term of the optical field and only intensity detection is required during the solving process. As a proof of principle, several 20 and 30-dimensional Ising problems have been solved with high ground state probability.
3 citations
12 Mar 2021
TL;DR: In this paper, a comparison of three types of phase shifters based on TiN metal, silicide and N-type doped silicon is presented, which were fabricated by the CUMEC SOI process on the same SOI die.
Abstract: The thermo-optic phase shifters (TOPS) have been widely used in the applications of sensing, lidar and neural network on the SOI platform. We present a comparison of TOPS based on TiN metal, silicide and N-type doped silicon, which were fabricated by the CUMEC SOI process on the same SOI die. The average switching power (Pπ) of these TOPS are19.12 mW, 21.75 mW and 21.96 mW, respectively. In addition, the switching time of these three types of TOPS have been tested under 10 KHz square wave, the rise and drop time are 5.90 μs and 8.97 μs, 12.90μs and 4.00 μs, 12.80μs and 2.60μs, respectively. Moreover, the minimum possible distance between adjacent TOPS was also examined, which is beneficial to the application of the TOPS in large-scale compact network.
3 citations
Cites background from "Deep learning with coherent nanopho..."
...While these devices are mainly used for telecommunication and data-interconnect industry, new applications such as optical phased array [4], photonic neural networks [5] and programmable photonics [6] are attracting increased interests....
[...]
04 Mar 2019
TL;DR: A new automated logic synthesis algorithm based on And-Inverter Graphs (AIGs) for electro-optic computing is proposed, which serves as the core part of the arithmetic logical unit (ALU) and several new proposed logic gates are presented.
Abstract: As a tremendous amount of data is being created exponentially day by day, integrated optical computing starts to attract lots of attention recently due to the bottleneck in the continuation of Moore’s law. With the rapid development of micro/nano-scale optical devices, integrated photonics has shown its potential to satisfy the demand of computation with an ultracompact size, ultrafast speed, and ultralow power consumption. As one of the paradigms in optical computing, the electro-optic logic that combines the merits of photonics and electronics has made considerable progress in various fundamental logic gates. It therefore becomes very critical to develop an automated design method to synthesize these logic devices for large-scale optical computing circuits. In this paper, we propose a new automated logic synthesis algorithm based on And-Inverter Graphs (AIGs) for electro-optic computing. A comprehensive component library of electro-optic logic is summarized with several new proposed logic gates. As an example, a large-scale ripple-carry full adder which serves as the core part of the arithmetic logical unit (ALU) is presented. In the design, all the electrical signals could be applied simultaneously at every clock cycle and then the light could process the signals through every bit at the speed of light without any delay accumulated. High-speed experiment demonstrations are carried out, which show its potential in future high-speed and low-power-consumption optical computing.
3 citations
References
More filters
Proceedings Article•
03 Dec 2012TL;DR: The state-of-the-art performance of CNNs was achieved by Deep Convolutional Neural Networks (DCNNs) as discussed by the authors, which consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax.
Abstract: We trained a large, deep convolutional neural network to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes. On the test data, we achieved top-1 and top-5 error rates of 37.5% and 17.0% which is considerably better than the previous state-of-the-art. The neural network, which has 60 million parameters and 650,000 neurons, consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax. To make training faster, we used non-saturating neurons and a very efficient GPU implementation of the convolution operation. To reduce overriding in the fully-connected layers we employed a recently-developed regularization method called "dropout" that proved to be very effective. We also entered a variant of this model in the ILSVRC-2012 competition and achieved a winning top-5 test error rate of 15.3%, compared to 26.2% achieved by the second-best entry.
73,978 citations
[...]
TL;DR: Deep learning is making major advances in solving problems that have resisted the best attempts of the artificial intelligence community for many years, and will have many more successes in the near future because it requires very little engineering by hand and can easily take advantage of increases in the amount of available computation and data.
Abstract: Deep learning allows computational models that are composed of multiple processing layers to learn representations of data with multiple levels of abstraction. These methods have dramatically improved the state-of-the-art in speech recognition, visual object recognition, object detection and many other domains such as drug discovery and genomics. Deep learning discovers intricate structure in large data sets by using the backpropagation algorithm to indicate how a machine should change its internal parameters that are used to compute the representation in each layer from the representation in the previous layer. Deep convolutional nets have brought about breakthroughs in processing images, video, speech and audio, whereas recurrent nets have shone light on sequential data such as text and speech.
46,982 citations
TL;DR: This work bridges the divide between high-dimensional sensory inputs and actions, resulting in the first artificial agent that is capable of learning to excel at a diverse array of challenging tasks.
Abstract: The theory of reinforcement learning provides a normative account, deeply rooted in psychological and neuroscientific perspectives on animal behaviour, of how agents may optimize their control of an environment. To use reinforcement learning successfully in situations approaching real-world complexity, however, agents are confronted with a difficult task: they must derive efficient representations of the environment from high-dimensional sensory inputs, and use these to generalize past experience to new situations. Remarkably, humans and other animals seem to solve this problem through a harmonious combination of reinforcement learning and hierarchical sensory processing systems, the former evidenced by a wealth of neural data revealing notable parallels between the phasic signals emitted by dopaminergic neurons and temporal difference reinforcement learning algorithms. While reinforcement learning agents have achieved some successes in a variety of domains, their applicability has previously been limited to domains in which useful features can be handcrafted, or to domains with fully observed, low-dimensional state spaces. Here we use recent advances in training deep neural networks to develop a novel artificial agent, termed a deep Q-network, that can learn successful policies directly from high-dimensional sensory inputs using end-to-end reinforcement learning. We tested this agent on the challenging domain of classic Atari 2600 games. We demonstrate that the deep Q-network agent, receiving only the pixels and the game score as inputs, was able to surpass the performance of all previous algorithms and achieve a level comparable to that of a professional human games tester across a set of 49 games, using the same algorithm, network architecture and hyperparameters. This work bridges the divide between high-dimensional sensory inputs and actions, resulting in the first artificial agent that is capable of learning to excel at a diverse array of challenging tasks.
23,074 citations
"Deep learning with coherent nanopho..." refers background or methods in this paper
...The computational resolution of ONNs is limited by practical non-idealities, including (1) thermal crosstalk between phase shifters in interferometers, (2) optical coupling drift, (3) the finite precision with which an optical phase can be set (16 bits in our case), (4) photodetection noise and (5) finite photodetection dynamic range (30 dB in our case)....
[...]
...(3) Once a neural network is trained, the architecture can be passive, and computation on the optical signals will be performed without additional energy input....
[...]
...We used four instances of the OIU to realize the following matrix transformations in the spatial-mode basis: (1) U((1))Σ((1)), (2) V((1)), (3) U((2))Σ((2)) and (4) V((2))....
[...]
...Transformations (1) and (2) realize the first matrix M((1)), and (3) and (4) implement M((2))....
[...]
TL;DR: In this article, an effective way of initializing the weights that allows deep autoencoder networks to learn low-dimensional codes that work much better than principal components analysis as a tool to reduce the dimensionality of data is described.
Abstract: High-dimensional data can be converted to low-dimensional codes by training a multilayer neural network with a small central layer to reconstruct high-dimensional input vectors. Gradient descent can be used for fine-tuning the weights in such "autoencoder" networks, but this works well only if the initial weights are close to a good solution. We describe an effective way of initializing the weights that allows deep autoencoder networks to learn low-dimensional codes that work much better than principal components analysis as a tool to reduce the dimensionality of data.
16,717 citations
TL;DR: This historical survey compactly summarizes relevant work, much of it from the previous millennium, review deep supervised learning, unsupervised learning, reinforcement learning & evolutionary computation, and indirect search for short programs encoding deep and large networks.
14,635 citations
"Deep learning with coherent nanopho..." refers methods in this paper
...ANNs can be trained by feeding training data into the input layer and then computing the output by forward propagation; weighting parameters in each matrix are subsequently optimized using back propagation [16]....
[...]