Deep learning with coherent nanophotonic circuits
Yichen Shen,Nicholas C. Harris,Scott Skirlo,Dirk Englund,Marin Soljacic +4 more
- Vol. 11, Iss: 7, pp 441-446
TLDR
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.read more
Citations
More filters
Journal ArticleDOI
Modeling and Analysis of Optical Modulators Based on Free-Carrier Plasma Dispersion Effect
Xuanqi Chen,Zhifei Wang,Yi-Shing Chang,Jiang Xu,Jun Feng,Peng Yang,Zhehui Wang,Luan H. K. Duong +7 more
TL;DR: A SPICE-compatible electro-optical co-simulation model, basic optical switch integration model (BOSIM), to systematically study optical modulators using PN, PIN, and metal–insulator–silicon (MIS) capacitor device technologies is proposed.
Journal ArticleDOI
Integrated Photonics Packaging: Challenges and Opportunities
Luigi Ranno,Parnika Gupta,Kamil Gradkowski,R. Bernson,Drew Weninger,Samuel Serna,Anuradha M. Agarwal,Lionel C. Kimerling,Juejun Hu,P. Obrien +9 more
TL;DR: In this article , the authors address the technical challenges and discuss promising strategies and research directions to overcome the "packaging bottleneck" in photonic integrated circuit (PIC) chips.
Journal ArticleDOI
Solving integral equations in free space with inverse-designed ultrathin optical metagratings
TL;DR: In this paper , an ultrathin Si metasurface-based platform for analogue computing that is able to solve Fredholm integral equations of the second kind using free-space visible radiation is presented.
Proceedings ArticleDOI
Optical Nonlinear Activation Functions Based on MZI-Structure for Optical Neural Networks
TL;DR: An on-chip optical nonlinear activation function circuit for optical neural networks based on a conventional linear transformer, MZI-mesh, which is reconfigurable to perform multiple types of non linear activation functions.
Proceedings ArticleDOI
Silicon Photonic Neuromorphic Computing with 16 GHz Input Data and Weight Update Line Rates
Apostolos Tsakyridis,George Giamougiannis,George Mourgias-Alexandris,Angelina Totovic,George Dabos,Nikolaos Passalis,Manos Kirtas,Anastasios Tefas,Miltiadis Moralis-Pegios,Nikos Pleros +9 more
TL;DR: A silicon photonic neuron is experimentally demonstrated that supports on-chip input-data and weight update rates at 16GHz and its computational performance is evaluated via the classification of the MNIST dataset achieving a mean accuracy of 99.18%.
References
More filters
Proceedings Article
ImageNet Classification with Deep Convolutional Neural Networks
TL;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.
Journal ArticleDOI
Deep learning
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.
Journal ArticleDOI
Human-level control through deep reinforcement learning
Volodymyr Mnih,Koray Kavukcuoglu,David Silver,Andrei Rusu,Joel Veness,Marc G. Bellemare,Alex Graves,Martin Riedmiller,Andreas K. Fidjeland,Georg Ostrovski,Stig Petersen,Charles Beattie,Amir Sadik,Ioannis Antonoglou,Helen King,Dharshan Kumaran,Daan Wierstra,Shane Legg,Demis Hassabis +18 more
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.
Journal ArticleDOI
Reducing the Dimensionality of Data with Neural Networks
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.
Journal ArticleDOI
Deep learning in neural networks
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.