All-optical nonlinear activation functions based on germanium silicon hybrid integration

2021 ◽  
Author(s):  
Hengkang Li ◽  
Jianji Dong
Keyword(s):  
Hybrid Integration ◽  
Activation Functions ◽  
All Optical ◽  
Germanium Silicon

scholarly journals All-optical neural network with nonlinear activation functions

Optica ◽  
2019 ◽  
Vol 6 (9) ◽  
pp. 1132 ◽  
Author(s):  
Ying Zuo ◽  
Bohan Li ◽  
Yujun Zhao ◽  
Yue Jiang ◽  
You-Chiuan Chen ◽  
...  
Keyword(s):  
Neural Network ◽  
Activation Functions ◽  
All Optical ◽  
Optical Neural Network

Ultrafast All-Optical Gating Operation Using Michelson Interferometer for Hybrid Integration of Intersubband Transition Switch on Si Platform

2011 ◽  
Vol 23 (24) ◽  
pp. 1884-1886 ◽  
Author(s):  
Masanori Nagase ◽  
Yuya Shoji ◽  
Satoshi Suda ◽  
Kenji Kintaka ◽  
Hitoshi Kawashima ◽  
...  
Keyword(s):  
Michelson Interferometer ◽  
Intersubband Transition ◽  
Hybrid Integration ◽  
Optical Gating ◽  
All Optical

Reconfigurable all-optical nonlinear activation functions for neuromorphic photonics

2020 ◽  
Vol 45 (17) ◽  
pp. 4819
Author(s):  
Aashu Jha ◽  
Chaoran Huang ◽  
Paul R. Prucnal
Keyword(s):  
Activation Functions ◽  
All Optical

Adaptive sigmoid-like and PReLU activation functions for all-optical perceptron

2021 ◽  
Author(s):  
Jasna Crnjanski ◽  
Marko Krstić ◽  
Angelina R. Totovic ◽  
Nikos Pleros ◽  
Dejan Gvozdic
Keyword(s):  
Activation Functions ◽  
All Optical

Correlation analysis of radiation damage

1978 ◽  
Vol 36 (1) ◽  
pp. 214-215
Author(s):  
R. Hegerl ◽  
A. Feltynowski ◽  
B. Grill
Keyword(s):  
Electron Microscopy ◽  
Independent Random Variables ◽  
Optical Parameters ◽  
Bright Field Image ◽  
Bright Field ◽  
Expectation Value ◽  
All Optical ◽  
Image Element ◽  
Stochastic Series ◽  
The Common

Till now correlation functions have been used in electron microscopy for two purposes: a) to find the common origin of two micrographs representing the same object, b) to check the optical parameters e. g. the focus. There is a third possibility of application, if all optical parameters are constant during a series of exposures. In this case all differences between the micrographs can only be caused by different noise distributions and by modifications of the object induced by radiation.Because of the electron noise, a discrete bright field image can be considered as a stochastic series Pm,where i denotes the number of the image and m (m = 1,.., M) the image element. Assuming a stable object, the expectation value of Pm would be Ηm for all images. The electron noise can be introduced by addition of stationary, mutual independent random variables nm with zero expectation and the variance. It is possible to treat the modifications of the object as a noise, too.

ALL-OPTICAL DIGITAL CIRCUITS

1988 ◽  
Vol 49 (C2) ◽  
pp. C2-459-C2-462 ◽  
Author(s):  
F. A.P. TOOLEY ◽  
B. S. WHERRETT ◽  
N. C. CRAFT ◽  
M. R. TAGHIZADEH ◽  
J. F. SNOWDON ◽  
...  
Keyword(s):  
Digital Circuits ◽  
All Optical

Comparison of various Activation Functions A Deep Learning Approach

2018 ◽  
Vol 6 (3) ◽  
pp. 122-126
Author(s):  
Mohammed Ibrahim Khan ◽  
◽  
Akansha Singh ◽  
Anand Handa ◽  
◽  
...  
Keyword(s):  
Deep Learning ◽  
Learning Approach ◽  
Activation Functions
Sign in / Sign up

Export Citation Format

Share Document