Вы находитесь на странице: 1из 36

Machine

Learning
Presentation by:

Karthik G
Final CSE
Dr. Mahalingam college of Engineering & technology,
Pollachi.
Code for Self Driving:

#include <stdio.h>
int main(){

if(“ Person in front of car”){


action(“Stop, don’t crush him”);
}

return 0;
}
Output:

Is there any
solution to solve
this problem?
Road Map

 What is Learning & Machine Learning


 Growth of ML
 Popular Algorithms
 Deep Learning
 Unsupervised Learning
 Reinforcement Learning
 Application of ML:
 Self Driving Cars
 Face Recognition
 Game Playing
 And much more…
What is“Learning”?
 Learning is the knowledge acquired through study,
experience, or being taught.

 It means understanding through data.

 Data is cheap and abundant (data warehouses, data


marts); knowledge is expensive and scarce.
What is Machine Learning?
 Machine Learning
 Study of algorithms that
 improve their performance
 at some task
 with “experience”

 It gives computers the ability to learn without being


explicitly programmed
- Arthur Samuel, 1959
Growth of Machine
Learning
1. Large Dataset (Big Data)
2. Powerful, Fast, Cheap Computers
(GPUs)
Earlier, only in large & expensive labs with less
computational power
Now, at our homes with faster processors
and internet
Applications & Algorithms of
Machine Learning
Learning Associations
 Basket analysis:
P (Y | X ) probability that somebody who buys X also buys Y
where X and Y are products/services.

Example: P ( bread | milk) = 0.7

Market-Basket transactions
TID Items
1 Bread, Milk
2 Bread, Diaper, Beer, Eggs
3 Milk, Diaper, Beer, Coke
4 Bread, Milk, Diaper, Beer
5 Bread, Milk, Diaper, Coke
Amazon Recommendation System
Classification (Logistic Regression)
 Example 1: Cancer
Classification
 Machine can learn
from size of tumor
which type of cancer it
is.

 Example 2: Spam Email Filtering


 Email System learn from the engagement
of the user whether the email is spam or
not.
Deep Learning
Artificial Neural Network Mimics
Human Brain

Human Brain Artificial Neural Network


How computer understands an image!!

Number One

Human Eye Machines


Deep Neural Network (Image Classification)

The more the hidden layers


Yes, Its the better the prediction or
Karthik!! feature detection
Facebook’s Auto-Tagging
Self Driving Cars!
LIDAR: Light Detection And Ranging

How cars see the surroundings?


Unsupervised Learning
 Learning “what normally happens”
 No output
 Clustering: Grouping similar instances
 Other applications: Summarization, Association Analysis
 Example applications
 Customer segmentation in CRM
 Image compression: Color quantization
 Bioinformatics: Learning motifs
Reinforcement Learning
 No supervised output but delayed reward
 Applications:
 Game playing
 Robot in a maze
 Multiple agents, partial observability, ...
Google’s AlphaGo Vs. Lee Sedol
Final Scores:
AlphaGo 4 vs.1 Lee Sedol

But why is it a big deal?


We play Chess with Computers too!!
Chess: Only 20 possible moves after each move
Go: Over 200 possible moves after each
move!!
Number of:

Atoms in Universe Possible moves in a Go game

10 80 10 170
Deep Q-Learning
Human Evolution:
Reference links:
1. https://aiexperiments.withgoogle.com
2. https://www.coursera.org/learn/machine-learning

Contact me:
E-mail: Riokarthy@gmail.com
Mobile: 915972273
Website: http://weeengineer.blogspot.com
As more and more artificial intelligence is entering into the
world, more and more emotional intelligence must enter
into leadership.
-Karthik.G

Thank you!

Вам также может понравиться