BASICS OF MACHINE LEARNING AND
ITS INDUSTRIAL IMPORTANCE
Mrs. Nagarathna C
Asst Prof. CSE, SKIT
FLOW OF PRESENTATION
 Definitions
Learning
Machine Learning (ML)
 When and Where to use ML
 Evolution of ML
 Learning Process
 Types of Learning
 Industry Application
 Job Opportunities
2 / 19
EXERCISE-1
WRITE AS MANY WORDS AS YOU CAN READ
ANSWER BELOW QUESTIONS
 How many words did you notice/see?
4/ 19
 How many words did you
write?
 How many words did you remember
more?
DEFINITIONS
 Learning: It is a process by which a system
improves performance from experience.
– Herbert Simon
5 / 19
 Machine Learning: Field of study that gives computers
the ability to learn without being explicitly programmed.
– Arther Samuel
 Machine Learning: It is the study of algorithms that
improves their performance <P>, at some task <T> with
experience <E>. A well defined learning task its given by
<P,T,E>
– Tom Mitchell
 Traditional Programming
COMPUTER
INPUT
PROGRAM OUTPUT
COMPUTER
 Machine Learning
INPUT
OUTPUT
PROGRAM
6/ 19
WHEN AND WHERE TO USE ML
1. Human expertise does not exist
Mars’ mission
2 . Human can’t explain their expertise
Speech recognition
3. Models can be customized
Medicines
4. Models are based on huge amount of data
genomics
7 / 19
EVOLUTION
Image: Linked In | Machine Learning vs Deep learning
8 / 19
KEYPOINTS TO REMEMBER
 Machine Learning works for monotonous task
 As knowledge tuned to specific task
 Is our brain multitask of
monotasked?
Hand rotation exercise
9 / 19
WHY ITS IS BEST TIME FOR ML
 System are now flooded with data
 Increase in computational power
 Growing progress is available algorithms and
theory developed by researchers
 Increasing acceptance and support from
industries
10 / 19
THE LEARNING PROCESS
11 / 19
Types of Learning
12 / 19
Supervised machine learning
13 / 19
Unsupervised machine learning
14 / 19
Reinforcement machine learning
In this, the machine learns from a hit and trial method.
Whenever the model predicts or produces a result, it is
penalized if the prediction is wrong or rewarded if the
prediction is correct. Based on these actions the model
trains itself.
15 / 19
Applications of Machine
learning
Virtual Personal Assistant
Facial recognition
E-mail spam filter
Recommendation engine on an e-commerce
website
16 / 19
Job Opportunities in ML
• Data Scientists
• Data Mining and Analysis
• Machine Learning Engineer
• AI Engineer
• Business Intelligence (BI) Developer
• Robotics programmer
17 / 19
18 / 19
“TorTure The daTa, and iT wilL confess To anyThing.”
– Ronald Coase
19/19

Ml introduction

  • 1.
    BASICS OF MACHINELEARNING AND ITS INDUSTRIAL IMPORTANCE Mrs. Nagarathna C Asst Prof. CSE, SKIT
  • 2.
    FLOW OF PRESENTATION Definitions Learning Machine Learning (ML)  When and Where to use ML  Evolution of ML  Learning Process  Types of Learning  Industry Application  Job Opportunities 2 / 19
  • 3.
    EXERCISE-1 WRITE AS MANYWORDS AS YOU CAN READ
  • 4.
    ANSWER BELOW QUESTIONS How many words did you notice/see? 4/ 19  How many words did you write?  How many words did you remember more?
  • 5.
    DEFINITIONS  Learning: Itis a process by which a system improves performance from experience. – Herbert Simon 5 / 19  Machine Learning: Field of study that gives computers the ability to learn without being explicitly programmed. – Arther Samuel  Machine Learning: It is the study of algorithms that improves their performance <P>, at some task <T> with experience <E>. A well defined learning task its given by <P,T,E> – Tom Mitchell
  • 6.
     Traditional Programming COMPUTER INPUT PROGRAMOUTPUT COMPUTER  Machine Learning INPUT OUTPUT PROGRAM 6/ 19
  • 7.
    WHEN AND WHERETO USE ML 1. Human expertise does not exist Mars’ mission 2 . Human can’t explain their expertise Speech recognition 3. Models can be customized Medicines 4. Models are based on huge amount of data genomics 7 / 19
  • 8.
    EVOLUTION Image: Linked In| Machine Learning vs Deep learning 8 / 19
  • 9.
    KEYPOINTS TO REMEMBER Machine Learning works for monotonous task  As knowledge tuned to specific task  Is our brain multitask of monotasked? Hand rotation exercise 9 / 19
  • 10.
    WHY ITS ISBEST TIME FOR ML  System are now flooded with data  Increase in computational power  Growing progress is available algorithms and theory developed by researchers  Increasing acceptance and support from industries 10 / 19
  • 11.
  • 12.
  • 13.
  • 14.
  • 15.
    Reinforcement machine learning Inthis, the machine learns from a hit and trial method. Whenever the model predicts or produces a result, it is penalized if the prediction is wrong or rewarded if the prediction is correct. Based on these actions the model trains itself. 15 / 19
  • 16.
    Applications of Machine learning VirtualPersonal Assistant Facial recognition E-mail spam filter Recommendation engine on an e-commerce website 16 / 19
  • 17.
    Job Opportunities inML • Data Scientists • Data Mining and Analysis • Machine Learning Engineer • AI Engineer • Business Intelligence (BI) Developer • Robotics programmer 17 / 19
  • 18.
  • 19.
    “TorTure The daTa,and iT wilL confess To anyThing.” – Ronald Coase 19/19