0% found this document useful (0 votes)
134 views29 pages

ANN (Perceptron and Multilayerd Perceptron)

The document discusses artificial neural networks and perceptrons. It describes how artificial neural networks are modeled after biological neural networks in the brain and are composed of interconnected nodes and weights that can learn patterns from data. The key components of a basic perceptron, which is a single artificial neuron, are described as the input layer, weighted sum, activation function, and output. Learning rules for updating the weights in a perceptron are also covered.

Uploaded by

ANMOL SINGH
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
134 views29 pages

ANN (Perceptron and Multilayerd Perceptron)

The document discusses artificial neural networks and perceptrons. It describes how artificial neural networks are modeled after biological neural networks in the brain and are composed of interconnected nodes and weights that can learn patterns from data. The key components of a basic perceptron, which is a single artificial neuron, are described as the input layer, weighted sum, activation function, and output. Learning rules for updating the weights in a perceptron are also covered.

Uploaded by

ANMOL SINGH
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 29

United college of engineering and research

Department of Computer Science and


Engineering
Machine Learning Technique
Presented By
Dr. Anita
.
OUTLINE

 ARTIFICIAL NEURAL
NETWORKS
– Perceptions'
– Multilayer perceptron, Gradient
– descent and the Delta rule,

2
INTRODUCTION
ARTIFICIAL NEURAL NETWORKS (ANN)

 Neural networks reflect the behavior of the human


brain, allowing computer programs to recognize
patterns and solve common problems in the fields
of AI, machine learning, and deep learning.

 Neural networks, also known as artificial neural


networks (ANNs) or simulated neural networks
(SNNs), are a subset of machine learning and are at
the heart of deep learning algorithms.

3
 Their name and structure are inspired by the
human brain, mimicking the way that biological
neurons signal to one another.

 The term "Artificial neural network" refers to a


biologically inspired sub-field of artificial
intelligence modeled after the brain.

 An Artificial neural network is usually a


computational network based on biological neural
networks that construct the structure of the human
4
brain
 Similar to a human brain has neurons
interconnected to each other, artificial neural
networks also have neurons that are linked to each
other in various layers of the networks.
 These neurons are known as nodes.

Figure. Biological Neural Network


5

Artificial Neural Network


 Dendrites from Biological Neural Network
represent inputs in Artificial Neural Networks, cell
nucleus represents Nodes, synapse represents
Weights, and Axon represents Output.
6
 There are around 1000 billion neurons in the
human brain.
 Each neuron has an association point somewhere
in the range of 1,000 and 100,000.

7
 consider an example of a digital logic gate that
takes an input and gives an output.
 "OR" gate, which takes two inputs.
 If one or both the inputs are "On," then we get
"On" in output.
 If both the inputs are "Off," then we get "Off" in
output.
 Here the output depends upon input.
 Our brain does not perform the same task.
 The outputs to inputs relationship keep changing
because of the neurons in our brain, which are
"learning." 8
 The architecture of an artificial neural
network:
 In order to define a neural network that consists of
a large number of artificial neurons, which are
termed units arranged in a sequence of layers.
 Artificial Neural Network primarily consists of
three layers:

9
 Input Layer:
 As the name suggests, it accepts inputs in several
different formats provided by the programmer.
 Hidden Layer:
 The hidden layer presents in-between input and
output layers. It performs all the calculations to
find hidden features and patterns.
 Output Layer:
 The input goes through a series of transformations
using the hidden layer, which finally results in
output that is conveyed using this layer.
10
 The artificial neural network takes input and
computes the weighted sum of the inputs and
includes a bias.
 This computation is represented in the form of a
transfer function.

 It determines weighted total is passed as an input


to an activation function to produce the output.
Activation functions choose whether a node should
fire or not.
11
12
Perceptron
 Perceptron was introduced by Frank Rosenblatt in
1957.
 Supervised Learning Algorithm
 used for Binary Classification
 There are two types of Perceptrons: Single layer
and Multilayer.
 Single layer - Single layer perceptrons can learn
only linearly separable patterns
 Multilayer - Multilayer perceptrons or feedforward
neural networks with two or more layers have the
greater processing power
13

 The Perceptron algorithm learns the weights for


the input signals in order to draw a linear decision
boundary.

Perceptron Learning Rule 14

 The Perceptron receives multiple input signals, a


 if the sum of the input signals exceeds a certain
threshold, it either outputs a signal or does not
return an output. In the context of supervised
learning and classification, this can then be used
to predict the class of a sample.
 Next up, let us focus on the perceptron function. 15
 Perceptron Function
 Perceptron is a function that maps its input “x,”
which is multiplied with the learned weight
coefficient; an output value ”f(x)”is generated.

 In the equation given above:


 “w” = vector of real-valued weights
 “b” = bias (an element that adjusts the boundary
away from origin without any dependence on the
input value)
 “x” = vector of input x values
16

 “m” = number of inputs to the Perceptron


 The output can be represented as “1” or “0.” It
can also be represented as “1” or “-1” depending
on which activation function is used.
 Inputs of a Perceptron 17

 A Perceptron accepts inputs, moderates them with


certain weight values, then applies the
transformation function to output the final result.

 A Boolean output is based on inputs such as


salaried, married, age, past credit profile, etc.

 It has only two values: Yes and No or True and


False. The summation function “∑” multiplies all
inputs of “x” by weights “w” and then adds them
up as follows:
18
19
 Activation Functions of Perceptron
 Activation function is internal state of neuron
 used to convert the input signal on node of ANN
to an output signal
 The activation function applies a step rule (convert
the numerical output into +1 or -1) to check if the
output of the weighting function is greater than
zero or not.
20

 For example:
 If ∑ wixi> 0 => then final output “o” = 1
 Else, final output -1
 Step function gets triggered above a certain value
of the neuron output; else it outputs zero.
 Sign Function outputs +1 or -1 depending on
whether neuron output is greater than zero or not.
 Sigmoid is the S-curve and outputs a value
between 0 and 1
21
 Output of Perceptron
 Perceptron with a Boolean output:
 Inputs: x1…xn
 Output: o(x1….xn)

 Weights: wi=> contribution of input xi to the


Perceptron output;
 w0=> bias or threshold
 If ∑w.x > 0, output is +1, else -1. The neuron gets
triggered only when weighted input reaches a
certain threshold value.

22

 An output of +1 specifies that the neuron


is triggered. An output of -1 specifies that
the neuron did not get triggered.
 “sgn” stands for sign function with output
+1 or -1.
 Multi-Layer Perceptrons (Feed forward Neural
Network)
 The field of artificial neural networks is often just
called neural networks or multi-layer perceptrons
after perhaps the most useful type of neural
network.
 A perceptron is a single neuron model that was a
precursor to larger neural networks.

23
 How do artificial neural networks work?
 Artificial Neural Network can be best represented
as a weighted directed graph, where the artificial
neurons form the nodes.
 The association between the neurons outputs and
neuron inputs can be viewed as the directed edges
with weights.
 The Artificial Neural Network receives the input
signal from the external source in the form of a
pattern and image in the form of a vector.
 These inputs are then mathematically assigned by
the notations x(n) for every n number of inputs.
24
Perceptron

25
 Afterward, each of the input is multiplied by its
corresponding weights ( these weights are the
details utilized by the artificial neural networks to
solve a specific problem ).
 In general terms, these weights normally represent
the strength of the interconnection between
neurons inside the artificial neural network.
 All the weighted inputs are summarized inside the
computing unit.

26
 If the weighted sum is equal to zero, then bias is
added to make the output non-zero or something
else to scale up to the system's response.
 Bias has the same input, and weight equals to 1.
Here the total of weighted inputs can be in the
range of 0 to positive infinity.
 Here, to keep the response in the limits of the
desired value, a certain maximum value is
benchmarked, and the total of weighted inputs is
passed through the activation function

27
 The activation function refers to the set of transfer
functions used to achieve the desired output.
 There is a different kind of the activation function,
but primarily either linear or non-linear sets of
functions.
 Some of the commonly used sets of activation
functions are the Binary, linear, and Tan
hyperbolic sigmoidal activation functions.

28
THANK YOU!

29

You might also like