Machine learning and its importance,
Industrial importance of Machine Learning
Industrial revolution 4.0
ML and its application
Python, Structured, semi-structured, unstructured
Evolution of Machine Learning
DC MACHINE-Motoring and generation, Armature circuit equation
Machine Learning(ML) and its industrial importance
1. BASICS OF MACHINE LEARNING
AND ITS INDUSTRIAL IMPORTANCE
Prof. Akhilesh Ladha
Lecturer, Information Technology,
R. C. Technical Institute,
Sola, Ahmedabad
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
Some Paradoxes
2 / 23
7. DEFINITIONS
Learning: It is a process by which a system
improves performance from experience.
– Herbert Simon
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
9. 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
10. ML NOT A GOOD OPTION
Payroll system
Billing system
GST calculation
University CPI/SPI calculation
and many more...
12. KEYPOINTS TO REMEMBER
Machine Learning works for monotonous task
As knowledge tuned to specific task
Is our brain multitask of
monotasked?
Hand rotation exercise
13. 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
19. 19
PARADOXES AND DILLEMAS
The Oceam Dillema: In Machine Learning,
accuracy and simplicity (i.e. Interpretability) are
in conflicts
LESS IS MORE
Learning from a high-dimensional feature space
requires an enormous amount of training to ensure
that there are several samples with each combination
of values.
With fixed number of training instances, the
predictive power reduces as the dimensionality
increases. /23
20. BONFERRONI’S PRINCIPLE
If you look in more places for interesting pattern
than your amount of data will support, you are
bound to find crap.
RHINE PARADOX
One should not tell other about there ESP
(Extra-Sensory Perception), that will cause them
to lose it.
20 / 23
21. R
ASHOMON
EFFECT
The Rashomon effect refers to an
instance when the same event is
described in significantly different
(often contradictory) ways by
different people who were involved.
21 / 23