
	Where results make sense

Topic: Backpropagation

Related Topics

Supervised learning

In the News (Thu 16 Feb 12)

EXECUTIVE POWER

Iran

ANALYSIS

Pakistan

Family compact

	Error Backpropagation
		This unsolved question was in fact the reason why neural networks fell out of favor after an initial period of high popularity in the 1950s.
		It took 30 years before the error backpropagation (or in short: backprop) algorithm popularized a way to train hidden units, leading to a new wave of neural network research and applications.
		Backpropagate the error: for l = L-1, L-2,..., 1,
	www.willamette.edu /~gorr/classes/cs449/backprop.html (785 words)

	Backpropagation (Neural Network Toolbox)
		A discussion of the architecture, simulation, and training of backpropagation networks
		A discussion of several high-performance backpropagation training algorithms
		A comparison of the memory and speed of different backpropagation training algorithms
	www.mathworks.com /access/helpdesk/help/toolbox/nnet/backprop.html (131 words)

	generation5 - Multi-Backpropagation Network: Concept and Modeling
		Backpropagation network is able to deal with various types of data and also has the ability to model a complex decision system.
		Backpropagation network with four input units and two hidden units for example required certain epochs, to create classification or prediction model.
		Backpropagation network is one of the well known NN model.
	www.generation5.org /content/2004/MultiBP.asp?Print=1 (869 words)

	Backpropagation Information
		Backpropagation requires that the transfer function used by the artificial neurons (or "nodes") be differentiable.
		Backpropagation usually allows quick convergence on satisfactory local minima for error in the kind of networks to which it is suited.
		The backpropagation algorithm for calculating a gradient has been rediscovered a number of times, and is a special case of a more general technique called automatic differentiation in the reverse accumulation mode.
	www.bookrags.com /wiki/Backpropagation (359 words)

	Causal Backpropagation
		Backpropagation is the mean by which an artificial neural network can learn through punitions resulting from an erroneous answer.
		The causal backpropagation component of such an network is external (although not independant) to the learning mechanism of the layers.
		In a sense, there is a perpetual competition between the temporal patterns the network recognizes so as to backpropagate and amplify causal signals, and the loss experienced by those same causal signals inherent to their passage inside the network.
	www.anticipation.info /texte/gastellu4/cbp.html (4184 words)

	The Generalized Delta Rule
		The main advantage of backpropagation over traditional methods of error minimization is that it reduces the cost of computing derivatives by a factor of N, where N is the number of derivatives to be calculated.
		Backpropagation is a procedure for efficiently calculating the derivatives of some output quantity of a nonlinear system, with respect to all inputs and parameters of that system, through calculations proceeding backwards from outputs to inputs.
		Backpropagation is any technique for adapting the weights of parameters of a nonlinear system by somehow using such derivatives or the equivalent.
	neuron-ai.tuke.sk /NCS/VOL1/P3_html/node31.html (1487 words)

	Backpropagation (Site not responding. Last check: )
		Static backpropagation is used to produce an instantaneous mapping of a static (time independent) input to a static output.
		At the core of all backpropagation methods is an application of the chain rule for ordered partial derivatives to calculate the sensitivity that a cost function has with respect to the internal states and weights of a network.
		In other words, the term backpropagation is used to imply a backward pass of error to each internal node within the network, which is then used to calculate weight gradients for that node.
	www.nd.com /definitions/backprop.htm (129 words)

	Using Backpropagation Network for the Estimation of Aqueous Activity Coefficients
		A simple Backpropagation network of nine input nodes (for the group coefficients), 11 hidden nodes, and one output node (for the activity coefficients) was developed and was compared with a simple regression model for its predicting power.
		Backpropagation networks, which are based on fully connected, layered, feedforward networks, in particular, have demonstrated the desirable properties of self-learning, noise-tolerance, and good predicting power.
		A Backpropagation network, as shown graphically in Figure 2, is a fully connected, layered, feedforward neural network.
	ai.bpa.arizona.edu /papers/sol92/sol92.html (4705 words)

	Dr. Dobb's \| The Backpropagation Neural Network \| April 15, 2003
		Neural networks almost died in the 1970s, but were reborn in the 1980s with the introduction of the backpropagation network.
		The backpropagation neural network has several layers of nodes, unlike the Adaline, which had only one node, and the Madaline which had one layer of nodes.
		Backpropagation's weight adjustment (training) process is somewhat involved and difficult to explain.
	www.ddj.com /184403122?pgno=14 (2278 words)

	Doctor AI: Progress Report
		The Backpropagation network is built on the same structure as the Multi-Layer Perceptron, but employs the backpropagation learning algorithm to train the network.
		A robust Backpropagation network was developed in Java based on backpropagation theory and several sample neural networks for the project.
		Backpropagation learning uses a computed output error that propagates from the output layer to the input layer, altering the weight values in the weight matrices.
	pages.cpsc.ucalgary.ca /~carman/533/progress (3262 words)

	Neural Network for Recognition of Handwritten Digits - The Code Project - Libraries & Projects
		Backpropagation is an iterative process that starts with the last layer and moves backwards through the layers until the first layer is reached.
		Backpropagation gives us a way to determine the error in the output of a prior layer given the output of a current layer.
		For one-threaded backpropagation the speed was around 12 patterns per second, whereas for two-threaded backpropagation the speed was around 23 patterns per second.
	www.codeproject.com /library/NeuralNetRecognition.asp (12822 words)

	Expert Prediction, Symbolic Learning, and Neural Networks: An Experiment on Greyhound Racing
		A Backpropagation network, as shown in Figure 1, is a multiple-layered, feed-forward neural network.
		In addition, the computing time needed for a Backpropagation network is directly proportional (with a factor often much greater than 1) to the number of hidden units and the number of hidden layers in a network.
		Similar to ID3 results, the Backpropagation algorithm obtained high payoffs for several long shots, i.e., $78 for race 10, $30.2 for race 37, $30 for race 64, and $34.8 for race 83.
	ai.bpa.arizona.edu /papers/dog93/dog93.html (6253 words)

	Introduction to Backpropagation Neural Networks (Site not responding. Last check: )
		The word Backpropagation stands for the second stage of the Neural Network learning process, when the system is adjusted, to make its output closer to what it should be.
		During the Backpropagation stage of the neural network training, the output layer of neurons is adjusted, to make the output closer to what it need to be.
		It is possible to write both supervised classification software, for example, using feedforward backpropagation algorythms, and not-superwised automatic classification, where the network will break the data into the cetegories.
	cortex.snowseed.com /neural_networks.htm (5135 words)

	Role of Dendritic Spines in Action Potential Backpropagation: A Numerical Simulation Study -- Tsay and Yuste 88 (5): ...
		The AP height ratio measured halfway along the apical dendrite (190 µm from soma) was used as measurement of backpropagation efficacy while varying channel densities and passive properties.
		A: backpropagation sensitivity to spine Na channel density indicated by plot of AP height ratio as measured halfway along the apical dendrite (190 µm from soma).
		Backpropagation has been shown to be influenced by dendritic morphology (Stuart et al.
	jn.physiology.org /cgi/content/full/88/5/2834 (6422 words)

	Backpropagation neural network C++ source code by Thomas Riga, University of Genoa, Italy, keywords: neural networks, ...
		Backpropagation networks are useful, among other tasks, for classification and generalization.
		This function is less complex to compute when a network is implemented on a digital computer than the sigmoid function, but it is not useful in a backpropagation algorhythm.
		They are the ideators of the backpropagation algorhythm: Parallel distributed processing: Explorations in the microstructure of cognition, J.L. McLelland, D.E. Rumelhart and the PDP research group, MIT press/Bradford Books, 1986
	members.tripod.com /zerkpage/backprop.htm (352 words)

	NeuroShell 2 (Site not responding. Last check: )
		This is the standard type of backpropagation network in which every layer is connected or linked only to the previous layer.
		This is the type of backpropagation network in which every layer is connected or linked to every previous layer.
		This type of backpropagation network is able to detect different features in the data with the use of multiple slabs of neurons in the hidden layer, each with a different activation function.
	www.wardsystems.com /products.asp?p=neuroshell2&page=2 (1131 words)

	The BackPropagation Network: Learning by Example (Draft)
		Since backpropagation uses a gradient-descent procedure, a BackProp network follows the contour of an error surface with weight updates moving it in the direction of steepest descent.
		The Backpropagation algorithm developed in this chapter only requires that the weight changes be proportional to the derivative of the error.
		Applied to backpropagation, the concept of momentum is that previous changes in the weights should influence the current direction of movement in weight space.
	www.cs.indiana.edu /~port/brainwave.doc/BackProp.html (6350 words)

	Backpropagation (Neural Network Toolbox)
		The backpropagation computation is derived using the chain rule of calculus and is described in Chapter 11 of [HDB96].
		The basic backpropagation training algorithm, in which the weights are moved in the direction of the negative gradient, is described in the next section.
		Momentum can be added to backpropagation learning by making weight changes equal to the sum of a fraction of the last weight change and the new change suggested by the backpropagation rule.
	www.ee.uwa.edu.au /~roberto/teach/matlab/toolbox/nnet/backpr54.html (1183 words)

	BACKPROPAGATION (Site not responding. Last check: )
		Backpropagation is the most popular, widely used implementation of an ANN.
		This network is fully connected, meaning that every neuron in level n-1 is connected not only to every other neuron in n-1, but also to every neuron in level n.
		First and foremost, backpropagation can be very slow because the training phase may take thousands or hundreds of thousands of iterations to find a minima associated with an acceptable level of error.
	www.ptproject.ilstu.edu /nn04.htm (140 words)

	Calcium-Dependent Persistent Facilitation of Spike Backpropagation in the CA1 Pyramidal Neurons -- Tsubokawa et al. 20 ...
		Persistent facilitation of spike backpropagation induced by depolarizing pulses.
		Tsubokawa H, Ross WN (1997) Muscarinic modulation of spike backpropagation in the apical dendrites of hippocampal CA1 pyramidal neurons.
		Tsubokawa H, Miura M, Kano M (1999a) Elevation of intracellular Na induced by hyperpolarization at the dendrites of pyramidal neurons of mouse hippocampus.
	www.jneurosci.org /cgi/content/full/20/13/4878 (4176 words)

	Character recognition
		Such a program could be useful when demonstrating how character recognition using backpropagation works or when demonstrating how many iterations of learning need to be performed to get a satisfactory result.
		Backpropagation is a technique discovered by Rumelhart, Hinton and Williams in 1986 and it is a supervised algorithm that learns by first computing the output using a feedforward network, then calculating the error signal and propagating the error backwards through the network.
		Many say that the slow speed of the backpropagation routine is a major drawback, but I consider this to be a minor drawback only since the speed of new computers doubles every third year and then speed becomes a less important issue than it is today.
	home.eunet.no /~khunn/papers/2039.html (1824 words)

	Backpropagation
		The backpropagation algorithm is a learning rule for multi-layered Neural Networks, credited to Rumelhart and McClelland.
		The Backpropagation algorithm is the most common network learning method.
		Backpropagation searches the space of possible hypotheses using gradient descent to iteratively reduce the error in the nuetwork fit to the training examples.
	www.cs.montana.edu /~grayd/backprop.htm (437 words)

	Herself's AI: Backpropagation Networks
		Backpropagation networks can be used on several network architectures.
		If you use a neural net that also accounts for imaginary numbers you can adapt this function so it is not always positive and calculate all of the four derivatives needed.
		Numerous iterations are required for a backpropagation network to learn.
	herselfsai.com /2007/03/backpropagation-networks.html (674 words)

	Recurrent Backpropagation (Site not responding. Last check: )
		When data is fed to the input of the network, the network cycles the data through the recurrent connections until it reaches a fixed output.
		Training a network using fixed-point learning can be more difficult than with static backpropagation, but the added power of these networks can result in much smaller and more efficient implementations.
		In recurrent backpropagation, activations are fed forward until a fixed value is achieved.
	www.nd.com /definitions/recback.htm (133 words)

	Backpropagation
		I believe matlab uses batch training for backpropagation which means that the weights and biases are updated after the whole training set has been applied to the net.
		This is important because there is nothing stopping the MATLAB user from programming their own training algorithm, whether it is backpropagation, L-M or something else entirely.
		Second, while L-M does a respectable job at training neural networks, and may be the fastest training algorithm for some problems, it is not "the fastest" for all problems.
	neuralnetworks.ai-depot.com /NeuralNetworks/1039.html (374 words)

	B/Q351: Delta Rule and Backpropagation
		Backpropagation: a gradient descent algorithm for learning the weights into hidden units as well as output units
		A network with hidden units with linear activation functions is equivalent to a network with no hidden units, at least the hidden units must have non-linear activation functions, and these must be differentiable for backpropagation to apply: usually the sigmoid function.
		Error is back-propagated, and weights are updated using the backpropagation rule, with context-to-hidden weights treated exactly as input-to-hidden weights.
	www.cs.indiana.edu /classes/b351-gass/Notes/backprop.html (956 words)

	Backpropagation with Momentum (Site not responding. Last check: )
		This algorithm is based on the basic backpropagation algorithm but introduces a momentum term.
		This tries to avoid oscillation problems common with the regular backpropagation algorithm when the error surface has a very narrow minimum.
		The net effect is that of traversing flat error surfaces quickly, while moving slower when the surface becomes irregular.
	helios.hampshire.edu /jdavila/proposal/with_momentum.htm (90 words)

	SAC'98 Paper: Fuzzy Clustering Improves Convergence of the Backpropagation Algorithm (Site not responding. Last check: )
		Since units in a network do not contribute equally to the overall error of the network, then their connections should be adjusted accordingly.
		So, it follows that a method that adjusts the connections between the units based on their contributions to the overall error of the network should yield an improved performance.
		This paper describes such a method by the use of fuzzy clustering of the error signals of the backpropagation algorithm in order to enhance the algorithm’s performance.
	www.cogsci.ed.ac.uk /~ceilidh/SAC-Papers/Paper51/index.html (304 words)

	Generation 5: Artificial Intelligence Repository - Multilayer Feedforward Network and the Backpropagation Algorithm
		The backpropagation algorithm is perhaps the most widely used training algorithm for multi-layered feedforward networks.
		However, many people find it quite difficult to construct multilayer feedforward networks and training algorithms from scratch, whether it be because of the difficulty of the math (which can seem misleading at first glance of all the derivations) or the difficulty involved with the actual coding of the network and training algorithm.
		Hopefully after you have read this guide you'll walk away knowing more about the backpropagation algorithm than you ever did before.
	library.thinkquest.org /18242/nn_bp.shtml (1517 words)

Try your search on: Qwika (all wikis)