You are on page 1of 12

ARTIFICIAL INTELLIGENCE

EL40102
CONTENTS

֍ Biological Neuron
֍ Artificial Neuron
֍ Activation Function
֍ Single Layer Perceptron
֍ Multi Layer Perceptron
֍ Back Propagation Algorithm
֍ Recurrent Neural Network
֍ Adaptive Resonance Theory

Asutosh Patnaik, Asst. Prof., Dept. of EEE, CGU,


10-08-2022 2
Odisha
Unit – 1 Lecture 3
Objective of Today’s Lecture
Understand the Neural Network Architecture and Activation Fuctions

Asutosh Patnaik, Asst. Prof., Dept. of EEE, CGU,


10-08-2022 3
Odisha
Artificial Neural Network
Architecture
 A neural network that consists of a large number of artificial neurons, which are termed units arranged in a
sequence of layers.
 ANN primarily consists of three layers.

 Input Layer:
 Accepts inputs.
 Hidden Layer:
 Presents in-between input and output
layers.
 It performs all the calculations to find
hidden features and patterns.

Asutosh Patnaik, Asst. Prof., Dept. of EEE, CGU,


10-08-2022 4
Odisha
Artificial Neural Network
Architecture
 A neural network that consists of a large number of artificial neurons, which are termed units arranged in a
sequence of layers.
 ANN primarily consists of three layers.

 Output Layer:
 Produces the output.

Asutosh Patnaik, Asst. Prof., Dept. of EEE, CGU,


10-08-2022 5
Odisha
Artificial Neural Network
Model

 x1, x2, x3, ……., xn are the n inputs to the artificial neuron.
 w1, w2, w3, ……., wn are the weights attached to the input links.

𝒚 = ∅(𝑰)
where ∅ is the activation function.
I y

Asutosh Patnaik, Asst. Prof., Dept. of EEE, CGU,


10-08-2022 6
Odisha
Artificial Neural Network
Example
Thresholding Function
𝒊 𝒏

𝒚=∅ 𝒘𝒊 𝒙𝒊 − 𝜽
𝒊 𝟏

Input
𝟏, 𝑰 > 𝟎
∅ 𝑰 =
𝟎, 𝑰 ≤ 𝟎

Threshold

Asutosh Patnaik, Asst. Prof., Dept. of EEE, CGU,


10-08-2022 7
Odisha
Artificial Neural Network
Example
Signum Function

+𝟏, 𝑰 > 𝟎
∅ 𝑰 = Input
−𝟏, 𝑰 ≤ 𝟎

Asutosh Patnaik, Asst. Prof., Dept. of EEE, CGU,


10-08-2022 8
Odisha
Artificial Neural Network
Example
Sigmoidal Function

𝟏
∅ 𝑰 = 𝜶𝑰
𝟏+𝒆

Asutosh Patnaik, Asst. Prof., Dept. of EEE, CGU,


10-08-2022 9
Odisha
Artificial Neural Network
Network Architecture

Single Layer Feedforward Networks

Asutosh Patnaik, Asst. Prof., Dept. of EEE, CGU,


10-08-2022 10
Odisha
Artificial Neural Network
Network Architecture

Multilayer Feedforward Networks

Asutosh Patnaik, Asst. Prof., Dept. of EEE, CGU,


10-08-2022 11
Odisha
Artificial Neural Network
Network Architecture

Recurrent Networks

Asutosh Patnaik, Asst. Prof., Dept. of EEE, CGU,


10-08-2022 12
Odisha

You might also like