％ Train a two layer neural network with the Levenberg-Marquardt
％ method.
％
％ If desired, it is possible to use regularization by
％ weight decay. Also pruned (ie. not fully connected) networks can
％ be trained.
％
％ Given a set of corresponding input-output pairs and an initial
％ network,
％ [W1,W2,critvec,iteration,lambda]=marq(NetDef,W1,W2,PHI,Y,trparms)
％ trains the network with the Levenberg-Marquardt method.
％
％ The activation functions can be either linear or tanh. The
％ network architecture is defined by the matrix NetDef which
％ has two rows. The first row specifies the hidden layer and the
％ second row specifies the output layer.
Train a two layer neural network with a recursive prediction error
％ algorithm ("recursive Gauss-Newton"). Also pruned (i.e., not fully
％ connected) networks can be trained.
％
％ The activation functions can either be linear or tanh. The network
％ architecture is defined by the matrix NetDef , which has of two
％ rows. The first row specifies the hidden layer while the second
％ specifies the output layer.
In this demo, I use the EM algorithm with a Rauch-Tung-Striebel smoother and an M step, which I ve recently derived, to train a two-layer perceptron, so as to classify medical data (kindly provided by Steve Roberts and Will Penny from EE, Imperial College). The data and simulations are described in: Nando de Freitas, Mahesan Niranjan and Andrew Gee Nonlinear State Space Estimation with Neural Networks and the EM algorithm After downloading the file, type "tar -xf EMdemo.tar" to uncompress it. This creates the directory EMdemo containing the required m files. Go to this directory, load matlab5 and type "EMtremor". The figures will then show you the simulation results, including ROC curves, likelihood plots, decision boundaries with error bars, etc. WARNING: Do make sure that you monitor the log-likelihood and check that it is increasing. Due to numerical errors, it might show glitches for some data sets.
Single-layer neural networks can be trained using various learning algorithms. The best-known algorithms are the Adaline, Perceptron and Backpropagation algorithms for supervised learning. The first two are specific to single-layer neural networks while the third can be generalized to multi-layer perceptrons.
外國人開發的電磁時域有限差分方法工具包
Electromagnetic Finite-Difference Time-Domain (EmFDTD)
is a basic two-dimensional FDTD code developed at the
School of Electrical Engineering, Sharif University of
Technology.
This code has been written based on the standard
Yee s FDTD algorithm. Applications include propagation,
scattering, and diffraction of electromagnetic waves
in homogeneous and non-homogeneous isotropic media
for in-plane propagating waves. Negative permittivites
or permeabilities as well as dispersion is not included.
Zero, Periodic, and Perfectly Matched Layer boundary
conditions may be selectively applied to the solution
domain.
The program is best suited for study of propagation and
diffraction of electromagnetic waves in Photonic Crystal
structures.
EmFDTD is written in MATLAB language and has been
tested under MATLAB 5.0 and higher versions.
Performance Comparison of Two On-Demand Routing Protocols Depends on Traffic Density by Using UWB-IR as Pyhisical and MAC Layer at Outdoor Peer to Peer Sensor Network
This document provides practical, common guidelines for incorporating PCI Express interconnect
layouts onto Printed Circuit Boards (PCB) ranging from 4-layer desktop baseboard designs to 10-
layer or more server baseboard designs. Guidelines and constraints in this document are intended
for use on both baseboard and add-in card PCB designs. This includes interconnects between PCI
Express devices located on the same baseboard (chip-to-chip routing) and interconnects between
a PCI Express device located “down” on the baseboard and a device located “up” on an add-in
card attached through a connector.
This document is intended to cover all major components of the physical interconnect including
design guidelines for the PCB traces, vias and AC coupling capacitors, as well as add-in card
edge-finger and connector considerations. The intent of the guidelines and examples is to help
ensure that good high-speed signal design practices are used and that the timing/jitter and
loss/attenuation budgets can also be met from end-to-end across the PCI Express interconnect.
However, while general physical guidelines and suggestions are given, they may not necessarily
guarantee adequate performance of the interconnect for all layouts and implementations.
Therefore, designers should consider modeling and simulation of the interconnect in order to
ensure compliance to all applicable specifications.
The document is composed of two main sections. The first section provides an overview of
general topology and interconnect guidelines. The second section concentrates on physical layout
constraints where bulleted items at the beginning of a topic highlight important constraints, while
the narrative that follows offers additional insight.
