T
Tyler W. Hughes
Researcher at Stanford University
Publications - 49
Citations - 1941
Tyler W. Hughes is an academic researcher from Stanford University. The author has contributed to research in topics: Artificial neural network & Laser. The author has an hindex of 17, co-authored 46 publications receiving 1157 citations. Previous affiliations of Tyler W. Hughes include University of Michigan.
Papers
More filters
Journal ArticleDOI
Training of photonic neural networks through in situ backpropagation and gradient measurement
TL;DR: A protocol for training photonic neural networks based on adjoint methods by physically backpropagating an optical error signal and calculating the gradient of the network with respect to its tunable degrees of freedom.
Journal ArticleDOI
Wave physics as an analog recurrent neural network
TL;DR: In this article, the authors identify a mapping between the dynamics of wave physics and the computation in recurrent neural networks, which indicates that physical wave systems can be trained to learn complex features in temporal data, using standard training techniques for neural networks.
Journal ArticleDOI
Training of photonic neural networks through in situ backpropagation.
TL;DR: In this article, the photonic analogue of the backpropagation algorithm is derived for computing gradients of conventional neural networks, and these gradients may be obtained exactly by performing intensity measurements within the device.
Journal ArticleDOI
Adjoint Method and Inverse Design for Nonlinear Nanophotonic Devices
TL;DR: This work presents an extension of the adjoint method to modeling nonlinear devices in the frequency domain, with the nonlinear response directly included in the gradient computation, to devise compact photonic switches in a Kerr nonlinear material.
Journal ArticleDOI
Reprogrammable Electro-Optic Nonlinear Activation Functions for Optical Neural Networks
TL;DR: In this article, an electro-optic hardware platform for nonlinear activation functions in optical neural networks is introduced, which allows for complete nonlinear on-off contrast in transmission at relatively low optical power thresholds and eliminates the requirement of having additional optical sources between each of the layers of the network.