Scribd
Upload
4 views

Machine Learning

Uploaded by

Ashita Tiwary Roll no. 6
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
4 views

Machine Learning

Uploaded by

Ashita Tiwary Roll no. 6
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 17

Skill Development

Training Program
on
ML

Training for Engineering students in the field of Emerging Technologies.

Training Partner

Vinoba Bhave University


Hazaribagh, Jharkhand (India)

Run by IITians & Industry Experts Follow Us


Machine Learning
Let’s train your first model
Machine Learning - Definition

Definition 1: Machine learning is a subfield of artificial intelligence that uses


algorithms trained on data sets to create model that enable machines to perform
tasks that would otherwise only be possible for humans, such as categorizing
images or predicting price.

Definition 2: Field of study that gives computers the ability to learn without being
explicitly programmed.
AI : Ability of computer
or machine to behave like
a human or complete task
which is only possible by
the human or simply say
just create an intelligence
Machine

Venn Diagram
Examples of AI
AI

Machine Learning

Venn Diagram
Machine Learning
Field of study that gives computers the ability to learn without being explicitly programmed.

Input Program

normal approach

output
Machine Learning Approach
Input(feature)
output(label)

How inputs and outputs decide, what will be


Machine Learning the good The Program ?
Approach

New input Program Predicted Value


What is Program here ?
A machine learning program establishes a
relationship between input and output.

Here Input is features (X) and output is label or


target or prediction (y).

Y = f(x);
Y(Price of House) = size(x1) * 1000 + floors
count (x2) * 5000;
x
Y = ax1 + bx2 + k;

Y = ax + b
You have programm and input you can
generate value very easily
Machine Learning Approach

x y

Program
Find Best Line
Here, Line is program.

Y = ax +b (general equation of
line)

Which line or program is best ,


how we decide ?

Suppose i have five line.


Line is best which have total
error is minimum
Types of Machine Learning
Regression
Regression algorithms are used to
predict continuous values such as price,
Supervised ML salary, age, etc

Classification
Machine Learning Classification algorithms are used to predict discrete
values such as Male or Female, True or False, Spam or
Not Spam, etc.

Unsupervised ML
Complete Flow of ML
Deep Learning.
Sometimes Normal Machine Learning Model is not enough for making prediction and
Cluster, then that time Deep Learning or multilayer neural network comes to the
picture
Performance of DL
for most flavors of the old
generations of learning algorithms
… performance will plateau. …
deep learning … is the first class of
algorithms … that is scalable. …
performance just keeps getting
better as you feed them more data

Andrew NG
AI

Data Science
ML

Domain Deep Learning and Neural


Knowledge Network
Mathematics

Venn Diagram
Thank
You
Follow Us

iHUB DivyaSampark @ IIT Roorkee

EduVitae Services

Training Partner

Skill Development Program by iHUB DivyaSampark, IIT Roorkee in collaboration with DST, (India)

You might also like