chapter 5

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 13

DAT 21003

Artificial Intelligence
Artificial Neural Network
Artificial Neural Network
Introduction
 An artificial neural network (ANN) consists
of a pool of simple processing units which
communicate by sending signals to each
other over many weighted connections.
 ANN is either a hardware implementation
or a computer program which strives to
simulate the information processing
capabilities.
 ANN is a technique for solving problems by
constructing software that works like our
brains.
Artificial Neural Network
Introduction

 There are two basic reasons why programmers interested in


building artificial neural network (ANN):
 Technical viewpoint: Some problems such as
character recognition or the prediction of future
states of a system require massively parallel and
adaptive processing.
 Biological viewpoint: ANN can be used to
replicate and simulate components of the human
(or animal) brain, thereby giving us insight into
natural information processing.
Artificial Neural Network
Introduction

 The “building blocks” of neural networks are the neurons.


 In technical systems, we also refer to them as units or nodes.
 Basically, each neuron
 receives input from many other neurons.
 changes its internal state (activation) based on the current
input.
 sends one output signal to many other neurons, possibly
including its input neurons (recurrent network).
Artificial Neural Network
How do ANN work?

 ANN are typically composed of a


great number of interconnected
artificial neurons. The artificial
neurons are simplified models of
their biological counterparts.
 An artificial neuron is a
mathematical function based on a
model of biological neurons, where
each neuron takes inputs, weighs
them separately, sums them up and An artificial neuron is an
passes this sum through a nonlinear imitation of a human neuron
function to produce output.
Artificial Neural Network
How do ANN work?

 ANN is a technique for solving problems by constructing


software that works like our brains.
 There are two types of architecture for ANN. These types
focus on the functionality of ANN:
 Single Layer Perceptron
 Multi-Layer Perceptron (MLP)
Artificial Neural Network
Single Layer Perceptron
 It is one of the oldest and first introduced
neural networks.
 Generally used in the linearly based cases
for the machine learning problems.
 The computation of a single layer
perceptron is performed over the
calculation of sum of the input vector
each with the value multiplied by
corresponding element of vector of the
weights.
 The value which is displayed in the output
will be the input of an activation
function.
Artificial Neural Network
Basic Components of Perceptron

 The basic components of a perceptron are:


 Input Layer: Consists of one or more input neurons, which
receive input signals from the external world or from
other layers of the neural network.
 Weights: Each input neuron is associated with a weight,
which represents the strength of the connection between
the input neuron and the output neuron.
 Bias:A bias term is added to the input layer to provide
the perceptron with additional flexibility in modeling
complex patterns in the input data.
Artificial Neural Network
Basic Components of Perceptron
 The basic components of a perceptron are:
 Activation Function: The activation function determines the output of the
perceptron based on the weighted sum of the inputs and the bias term.
Common activation functions used in perceptron include the step
function, sigmoid function, and Rectified Linear Unit (ReLU) function.
 Output: The output of the perceptron is a single binary value, either 0 or
1, which indicates the class or category to which the input data belongs.
Artificial Neural Network
Multi-Layer Perceptron (MLP)
 MLP is a powerful member of the ANN family that can be used
to solve complex problems that a single perceptron alone
cannot solve.
 MLP are that kind of Feed-forward network, in which the data
is passed only in one direction.
 MLP is composed of multiple layers of interconnected
neurons.
Artificial Neural Network
Multi-Layer Perceptron (MLP)
 The structure of MLP can be broken down
into three main parts: the input layer, the
hidden layers, and the output layer.
 The input layer receives the input data
and passes it on to the next layer.
 The hidden layer contains a set of
interconnected neurons, which process
and analyze the input data passed on
from the previous layer.
 The output layer receives the output
from the previous layers, combines
them, and produces the final output
which is the solution to the problem.
Artificial Neural Network
Advantages and disadvantages of Multi-Layer Perceptron (MLP)
 Advantages:
 Can be applied to complex non-linear problems.
 Works well with big data.
 Provides quick predictions after training.
 The same accuracy ratio can be achieved even with smaller data.
 Disadvantages
 Itis not known to what extent each independent variable is affected by the
dependent variable.
 Computations are difficult and time consuming.
 The proper functioning of the model depends on the quality of the training
data. If the model does not work properly, generalization problems arise.
Artificial Neural Network
Factors affecting ANN performance
 Number of hidden nodes:
 Too many and the network may memorize training data
 Too few and the network may not learn the training data
 Initial weight set:
 some starting weight sets may lead to a local minimum
 other starting weight sets avoid the local minimum
 Training data:
 must be statistically relevant
 patterns should be presented in random order

You might also like