Scribd
Upload
64 views3 pages

Mahendra Engineering College: Lecture Handouts

The document provides an overview of the backpropagation algorithm. It begins with an introduction that defines backpropagation as a multilayered, feedforward neural network that is commonly used for supervised training. The document then discusses the prerequisites, training and testing phases, and operations of backpropagation networks. Specifically, it notes that backpropagation networks can have multiple hidden layers and involve a feedforward step to propagate input and an error backpropagation step to update weights. Finally, it provides a video and book reference for further learning.

Uploaded by

SHAIK ROSHAN
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
64 views3 pages

Mahendra Engineering College: Lecture Handouts

The document provides an overview of the backpropagation algorithm. It begins with an introduction that defines backpropagation as a multilayered, feedforward neural network that is commonly used for supervised training. The document then discusses the prerequisites, training and testing phases, and operations of backpropagation networks. Specifically, it notes that backpropagation networks can have multiple hidden layers and involve a feedforward step to propagate input and an error backpropagation step to update weights. Finally, it provides a video and book reference for further learning.

Uploaded by

SHAIK ROSHAN
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 3

L-4

MAHENDRA ENGINEERING COLLEGE


MAHENDHIRAPURI, MALLASAMUDRAM - 637503
Dept.EIE VI/III

Lecture Handouts
Subject Name : NEURAL NETWORK AND FUZZY LOGIC CONTROL

Staff Name : ABIRAMI.A AP/ECE

Unit : I (ARTIFICIAL NEURAL NETWORK)


Date of Lecture: 11.12.15

Topic of Lecture: Back propagation Algorithm(BPA)

Introduction :
 The Back propagation neural network is a multilayered, feed forward neural network and is
by far the most extensively used.
 It is also considered one of the simplest and most general methods used for supervised
training of multilayered neural networks.
 Back propagation works by approximating the non-linear relationship between the input and
the output by adjusting the weight values internally.
 It can further be generalized for the input that is not included in the training
patterns (predictive abilities).

Prerequisite knowledge for Complete learning of Topic:


Training patterns
Testing Patterns
Detailed content of the Lecture:
Back propagation network has two stages, training and testing.
 During the training phase, the network is "shown" sample inputs and the correct
classifications.
 For example, the input might be an encoded picture of a face, and the output could be
represented by a code that corresponds to the name of the person.

 A further note on encoding information - a neural network, as most learning algorithms, needs
to have the inputs and outputs encoded according to an arbitrary user defined scheme.

 The scheme will define the network architecture so that once a network is trained, the scheme
cannot be changed without creating a totally new net.

 Similarly there are many forms of encoding the network response.


 The following figure shows the topology of the Back propagation neural network that
includes and input layer, one hidden layer and an output layer.

 It should be noted that Back propagation neural networks can have more than one hidden
layer.
 The operations of the Back propagation neural networks can be divided into two steps: feed
forward and Back propagation
 In the feed forward step, an input pattern is applied to the input layer and its effect
propagates, layer by layer, through the network until an output is produced.
 The network's actual output value is then compared to the expected output, and an error signal
is computed for each of the output nodes.

Feed forward:
When a specified training pattern is fed to the input layer, the weighted sum of the input to the j th node
in the hidden layer is given by

The term is the weighted value from a bias node that always has an output value of 1.

Video Content / Details of website for further learning (if any):


https://www.youtube.com/watch?v=aVId8KMsdUU
Important Books/Journals for further learning including the page nos.:
Neural networks,Fuzzy logic and genetic algorithms by G.A.Vijayalakshmi Pai July 2012.
Pg:no:41-42.

Subject Teacher

Verified by HOD

You might also like