Scribd
Upload
14 views

Polymers

The document discusses neural networks, including what they are, their basic structure, objectives, properties, applications, and comparison to biological neural networks. A neural network is a machine learning method inspired by the human brain that uses interconnected nodes like neurons arranged in layers. They can learn from data, self-organize, operate in parallel in real time, and be fault tolerant. Neural networks are used for tasks like pattern recognition, prediction, control systems, and more.

Uploaded by

Bikash Mahaseth
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
14 views

Polymers

The document discusses neural networks, including what they are, their basic structure, objectives, properties, applications, and comparison to biological neural networks. A neural network is a machine learning method inspired by the human brain that uses interconnected nodes like neurons arranged in layers. They can learn from data, self-organize, operate in parallel in real time, and be fault tolerant. Neural networks are used for tasks like pattern recognition, prediction, control systems, and more.

Uploaded by

Bikash Mahaseth
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 20

NEURAL

NETWORKS
Shaina Gupta
22MMA001
CONTENTS:
◦ What is a Neural Network?
◦ Basic Neural Network
◦ Objectives of a Neural Network
◦ Properties of a Neural Network
◦ Applications of a Neural Network
◦ Biological and Artificial Neural Network
◦ Correlation between artificial and biological neuron
◦ Important terms and Parameters associated with Neural Network
◦ Some commonly used activation functions
WHAT IS A NEURAL NETWORK?

◦ A neural network is a method in artificial intelligence that


teaches computers to process data in a way that is inspired
by the human brain. It is a type of machine learning
process, called deep learning, that uses interconnected
nodes or neurons in a layered structure that resembles the
human brain.
BASIC NEURAL NETWORK
OBJECTIVES OF A NEURAL NETWORK:

◦ Learning/Training
◦ Generalization
◦ Application
PROPERTIES OF NEURAL NETWORK:

◦ ADAPTIVE LEARNING:
Neural Networks have the ability to learn how to do tasks based on the data given for training or
initial experience.

◦ SELF-ORGANISATION:
Neural Networks can create their own organization or representation of data during learning time.
◦ REAL TIME ORGANISATION:
Neural Networks have the ability to carry out computations in parallel in real time.

◦ FAULT TOLERANCE:
If a particular neuron or a group of neurons are damaged, even then some network capabilities are
retained because of it’s distributed architecture.
APPLICATIONS OF A NEURAL NETWORK:

◦ Control Systems
◦ Signal Processing
◦ Industrial Automation
◦ Image, Voice and Pattern Recognition
◦ Medical Imaging and Diagnosis
◦ Database Management
◦ Data Mining and Processing
◦ Weather Prediction and Forecast
◦ Traffic Control
◦ Economic and Business Modelling
BIOLOGICAL
AND
ARTIFICIAL
NEURAL
NETWORK
BIOLOGICAL NEURAL NETWORK

◦ Neural Networks are processing systems, the design of which is


inspired by the structure and functioning of the human brain.
◦ The human brain comprises of a large number of neurons. It has
approximately 10^11 neurons which are interconnected with each
other.
BIOLOGICAL NEURON
MAIN PARTS OF A BIOLOGICAL NEURON

◦ DENDRITES
They are tree like structural networks made up of nerve fibers. They are connected to soma or cell body.

◦ AXON
It is a single, long connection extending from the cell body. It carries signals from the neuron.

◦ SOMA OR CELL BODY


It is the main structural part of the neuron which carries the nucleus.
ARTIFICIAL NEURON
MAIN PARTS OF AN ARTIFICIAL NEURON

◦ INPUT

◦ WEIGHT ASSOCIATED WITH EACH INPUT

◦ ACTIVATION FUNCTION OF THE NEURON

◦ OUTPUT
CORRELATION BETWEEN BIOLOGICAL AND
ARTIFICIAL NEURON
BIOLOGICAL NEURON ARTIFICIAL NEURON

CELL NEURON/NODE

DENDRITES/SYNAPSE WEIGHTS

SOMA NET INPUT

AXON OUTPUT
IMPORTANT TERMS AND PARAMETERS
ASSOCIATED WITH NEURAL NETWORK
◦ INPUT NODES:
These are the nodes to which the input signals are applied.

◦ OUTPUT NODES:
These are the nodes from which the final output is taken.

◦ HIDDEN/INTERMEDIATE NODES:
These are the nodes that lie in between input and output node layers.
◦ WEIGHTS
Each neuron in the neural network is interconnected with other neurons by direct communication links.
Each link is associated with a weight, which contains information about the signal flowing through the
link.

◦ BIAS
The bias is a constant signal value which is added to a neuron. It is just another weighted link with a
constant signal value.
◦ ACTIVATION FUNCTION
Each neuron or node has an activation function associated with it and determines the input-output
relationship for that neuron. It can either be linear or non-linear.
SOME COMMONLY USED ACTIVATION
FUNCTIONS
◦ IDENTITY FUNCTION

◦ SIGMOIDAL FUNCTION
◦ BINARY STEP FUNCTION

◦ THRESHOLD

You might also like