Scribd
Upload
27 views

Neural Networks-Notes

Artificial Neural Networks (ANN), also called Neural Networks, are systems inspired by the human brain that can be used to solve complex computational problems. ANNs regulate the activation of computing cells similar to how neurons activate and deactivate in the human brain. A neural network consists of interconnected nodes, similar to neurons, that can adapt as they are exposed to new data. Neural networks are organized in layers including an input layer, hidden layers, and an output layer. Data moves through the network as it is processed to produce an output. Neural networks can learn through supervised, unsupervised, or reinforcement learning methods and are widely used for applications like computer vision, natural language processing, and more.

Uploaded by

MEGHA MENON
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
27 views

Neural Networks-Notes

Artificial Neural Networks (ANN), also called Neural Networks, are systems inspired by the human brain that can be used to solve complex computational problems. ANNs regulate the activation of computing cells similar to how neurons activate and deactivate in the human brain. A neural network consists of interconnected nodes, similar to neurons, that can adapt as they are exposed to new data. Neural networks are organized in layers including an input layer, hidden layers, and an output layer. Data moves through the network as it is processed to produce an output. Neural networks can learn through supervised, unsupervised, or reinforcement learning methods and are widely used for applications like computer vision, natural language processing, and more.

Uploaded by

MEGHA MENON
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/ 24

Artificial Neural Networks

Artificial Neural Networks(ANN) also called Neural Networks are artificially created
systems or networks that are based on the working of human brain . These networks solve
complex computation problems by regulating how the computing cells are activated,
similar to the activation and deactivation of the neurons in the human brain.
.

What is Neural Network ?


What is brain?

• Brain is the CPU for a human that


controls the whole body
• Brain is a huge and complex network
of approximately 100 billion neurons,
with each Neuron connects
approximately to 104 other neurons.
• The whole brain can make complex
decisions with the help of this
neurons.
Dendrite carry
information
from outside
to the cell.

The axon
carry
information
away from the
cell body.

STRUCTURE OF A NEURON
Mechanism Of Neurons
A neuron consists of a Soma( cell body ),
many dendrites and Axon
Pulses from other neurons are inputted to the
cell body from the dendrites via synapses and
are integrated there.
The neuron send a pulse to other neurons
through the Axon when the potential of the
Cell body is higher than the Threshold
Neural Networks

• It can extract the data features without any programming.

• The computer can learn, recognize and make decisions like human
beings.

• It works on machine learning algorithms to perform the task.

• It is a fast and efficient way to solve problems for which the dataset
is very large.
Let us learn the design
Layers of Neural networks

• Input Layer
• Hidden layer
• Output layer
ARTIFICIAL NEURON NETWORK WITH TWO
HIDDEN LAYERS
Working
Working

• This is a representation of how neural networks works.


• A neural network is divided into multiple layers, each layer
is further divided into several blocks called Nodes.
• Each node has its own task to accomplish which is then
passed to the next layer.
Layers of Neural Network

• The first layer of neural network is known as Input layer.


• The job of an input layer is to acquire data and feed it into the neural
network.
• No processing occurs in the input layer.
• Next to it are hidden layers.
• Hidden layers are layers in which the whole processing occurs.
• Their name essentially means that these layers are hidden and are not
visible to the user.
How it works

• Each layer of these hidden layers has its own machine learning
algorithm which it executes on the data received from the input
layer.
• The processed output is then fed into the subsequent hidden layer of
the network.
• There can be multiple hidden layers in a neural network system and
their number depends on the complexity of function for which the
network has been configured.
How it works

• Also the number of nodes in each layer can vary accordingly.


• The last hidden layer passes the final processed data to the output
layer , which then gives it to the user as the final output
• Similar to input layer output layer too does not process the data which it
acquires.
• It is meant for user interface.
Performance of Neural networks
How Machines learn

There are three ways of learning


1. Supervised learning
2. Unsupervised learning
3. Reinforcement learning
How Machines learn

SUPERVISED LEARNING
• Supervised learning uses labelled dataset to train the
model.
• Here the machine know the features of the object and
the labels associated with those features
How Machines learn

UNSUPERVISED LEARNING
• Unsupervised learning uses unlabelled dataset to
train the model.
• Here the model learns by extracting features and
patterns on its own.
How Machines learn

REINFORCEMENT LEARNING
• Reinforcement learning is learning based on reward or
punishment.
• It works on the basis of the feedback provided to the model.
Features of Neural Networks
Applications of neural networks

• Voice assistants like Siri,Alexa


• Language translators like Google Translate
• Automated vehicles like Tesla, Waymo
• Face recognition
• Automated music composition softwares like
Musenet
Quick draw(Neural network is behind this
game)

https://quickdraw.withgoogle.com/
This game will try to guess what you are drawing while you are drawing it. The game
will ask you to draw six drawings, one after the other.You will have 20 seconds to draw
each of the figures.
USING HANDWRITING WITH A NEURAL
NETWORK.

https://distill.pub/201
6/handwriting/
This is an experiment that uses neural networks for
generating strokes based on the inputs provided by you.
Multi predictor experiment

https://magenta.tensorflow.org/assets/sketch_rnn_demo/multi_predict.html
This experiment predicts the drawing which the user wants to create based on the
partial drawing made by the user.For using this experiment you start drawing a
sketch on the left side area of the screen.The neural network will predict the
complete drawing based on your inputs and show it on the right side of the
screen.The AI in this experiment will predict different outputs based on the
probabilities.

You might also like