Brief Introduction on Machine Learning and Deep Learning and it's some algorithms and also How to start career Machine Learning and Deep Learning.

1. Overview of Machine Learning & Applications
Mr. Deepak Chawla
Associate Software Developer at Tek
Palette, Bhilwara, Raj.
Email: deepakchawla35@gmail.com
WebLink: www.deepakchawla.me
Follow me on:-
Quora and Linkedin

2. Table of Content
■ Artificial Intelligence
■ Machine Learning
■ Why Machine Learning is Possible now…??
■ Working of Machine Learning
■ Types of Machine Learning
■ Neural Network or Deep Learning
■ Working of Neural Network
■ Some popular Neural Network Algorithms
■ Application of Machine Learning and Deep Learning
■ Machine Learning IT industry news
■ Machine Learning Engineering requirements and salary
■ Tools/Libraries used in Machine Learning
■ Prerequisite for Machine Learning
■ Key People to Follow
■ Resources and References
■ Q & A

3. What is Artificial intelligence (AI)?
❏ According to the father of
Artificial Intelligence, John
McCarthy, it is “The science and
engineering of making intelligent
machines, especially intelligent
computer programs”.
❏ Artificial Intelligence is a way of
making a computer, a computer-
controlled robot, or a software
think intelligently, in the similar
manner the intelligent humans
think.
3

4. Artificial Intelligence
■ Artificial Intelligence
◻ Ability of machines in conducting intelligent tasks
■ Intelligent Programs
◻ Programs conducting specific intelligent tasks
Input
Intelligent
Processing
Output
4

5. Data Science
5

6. A Few Quotes
■ “A breakthrough in machine learning would be worth ten
Microsoft's” (Bill Gates, Chairman, Microsoft)
■ “Machine learning is the next Internet” (Tony Tether, Director,
DARPA)
■ Machine learning is the hot new thing” (John Hennessy, President,
Stanford)
■ “Web rankings today are mostly a matter of machine
learning” (Prabhakar Raghavan, Dir. Research, Yahoo)
■ “Machine learning is going to result in a real revolution”
(Greg Papadopoulos, CTO, Sun)
6

7. What is Machine Learning?
(Machine Learning: Modern Approaches to Artificial Intelligence)
■ Definitions (Arthur Samuel, 1959)
◻Machine learning is the subfield of computer science that
gives computers the ability to learn without being explicitly
programmed
■ Definitions (Mitchell, 1997)
◻ A computer program is said to learn from experience E with
respect to some class of tasks T and performance measure
P, if its performance at the tasks improves with the
experiences
7

8. 8
Computer
Computer
Data
Program
Data
Output
Output
Program
Traditional Programming
Machine Learning

9. Why machine learning? Why now?
9

10. 1.Big Data Growth 2010-15 Per Month
10

11. Today’s Data Growth Per Minute
11

12. 2.Computation Power Growth
12

13. Prediction
Working of Machine Learning
Training
Labels
Training
Images
Training
Trainin
g
Image
Features
Image
Features
Testin
g
Test
Learned
model
Learned
model
13

14. Algorithms
Supervised learning Unsupervised learning
Semi-supervised learning 14

15. 15
Types of Machine Learning

16. ■ Supervised learning
Learning
16

17. ■ Supervised learning
◻In supervised learning, we are given a data set and already
know what our correct output should look like, having the idea
that there is a relationship between the input and the output.
◻Supervised learning of two types:
◻Regression Problem :-We are trying to predict results within a
continuous output.
◻Classification Problem :-We are instead trying to predict results
in a discrete output (yes or no).
Learning
17

18. ■ Our goal is predict output in continuous values.
■ So let’s say our simple predictor has this form:-
Linear Regression Problem
18
■ This function is known as Hypothesis Function.
■ To fit a best possible straight line we use a cost function
which knows as Cost Function.
■ To optimization of cost function we use Gradient
Descent Function which gives us global minimum
value.

19. Fig.1
19

20. Fig.2
20

21. Fig.3
21

22. Fig.4
22

23. Fig.5
23

24. Fig.6
24

25. ■ Unsupervised learning
Learning
25

26. ■ Unsupervised learning
◻Unsupervised learning allows us to approach problems with little
or no idea what our results should look like. We can derive
structure from data where we don't necessarily know the effect of
the variables.
◻We can derive this structure by clustering the data based on
relationships among the variables in the data.
◻With unsupervised learning there is no feedback based on the
prediction results.
Learning
26

27. ■ Reinforcement Learning
◻It is the problem of getting an agent to act in the world so as to
maximize its rewards.Consider teaching a dog a new trick: you
cannot tell it what to do, but you can reward/punish it if it does
the right/wrong thing. It has to figure out what it did that made it
get the reward/punishment, which is known as the credit
assignment problem.
Learning
27

28. Example 1: Disease Diagnosis
Disease
classifier
New patient’s
data
Presence or
absence
Database of medical records
Patient 1’s data Absence
Patient 2’s data Presence
… …
Training
28

29. Example 2: Chess Playing
Strategy of
Searching and
Evaluating
New matrix
representing
the current
board
Best move
Games played:
Game 1’s move list Win
Game 2’s move list Lose
… …
Training
29

30. Example 3: Text Classification
Text classifierNew text file class
Classified text files
Text file 1 trade
Text file 2 ship
… …
Training
30

31. 31
Neural Network or Deep Learning
■ Algorithms that try to mimic the
human brain.
■ Deep learning is a form of machine
learning that uses a model of
computing that's very much
inspired by the structure of the
brain. Hence we call this model a
neural network. The basic
foundational unit of a neural
network is the neuron, which is
actually conceptually quite simple.

32. Working of Human Brain
32

33. Learning Model of Neural Networks
33

34. Comparison Graph
34

35. Some Popular Deep Learning Algorithms
■ Feed-forward neural networks
■ Recurrent neural network
■ Multi-layer perceptrons (MLP)
■ Convolutional neural networks
■ Recursive neural networks
■ Deep belief networks
■ Convolutional deep belief networks
■ Self-Organizing Maps
■ Deep Boltzmann machines
■ Stacked de-noising auto-encoders
35

36. Alvinn Self Driving Car (1989)
36

37. Sample Applications
■ Web search
■ Computational biology
■ Finance
■ E-commerce
■ Space exploration
■ Robotics
■ Information extraction
■ Social networks
■ [Your favorite area]
37

38. Some News relates to AI
38
Source:- https://fossbytes.com/googles-ai-codes-own-machine-learning-software/

39. Some News relates to AI
39
Source:- http://economictimes.indiatimes.com/tech/ites/infosys-releases-9000-employees-due-to-
automation/articleshow/56680032.cms

40. Some News relates to AI
40
Source:- https://fossbytes.com/googles-ai-codes-own-machine-learning-software/

41. Machine Learning Jobs
41
Source:- http://www.payscale.com/research/IN/Skill=Machine_Learning/Salary

42. Tools/Libraries used in ML
■ Data Analysis
◻ Scikit Learn
◻ Hadoop or Spark Mllib
◻ Panda
◻ Matplotlib
■ Image Recognition
◻ OpenCV
◻ Dlib
◻ Tensor Flow
◻ Caffe
■ Speech and Voice Recognition
◻ Tensor Flow
◻ NLTP
◻ Kaldi
◻ CMU Sphinx
42

43. Prerequisite to start Machine Learning
■ Probability and Statistics Theory
■ Computer Science with Coding Skills.
■ 1-2 Experience in any domain(Web/Mobile/Desktop).
■ Advanced knowledge of :-
◻ Linux Operating System as a CLI not GUI
◻ Shell Scripting
◻ Networking
◻ C/C++/Java/Python/R
◻ DSA & DAA
◻ Cloud Computing
■ “Dedication, Curiosity, Patience & Self Confidence”
43

44. Resources to study ML
■ Google
■ Github
■ Youtube
■ Coursera
■ Stanford University Online Lectures
■ Carnegie Mellon University Online Lectures
■ Kdnuggets
■ Udacity
■ ML,AI,DL Groups on Linkedin, Facebook,Google Plus
■ Read Blogs & Research Papers on ML,AI,DL
■ Quora
44

45. Key People to Follow
Mark Zuckerberg, Founder of Facebook
45
Andrew Ng, is VP & Chief
Scientist of Baidu, Co-
Founder of Coursera and an
Adjunct Professor at Stanford
University.
Yann Lecun,
Director of AI Research, Facebook

46. Key People to Follow
46
Geoffrey Hinton, a researcher at Google
and professor emeritus at the
University of Toronto.Also known by
Godfather of Deep Learning.
Fei Fei Li Associate Professor at the
C.S. Department at Stanford
University and Also known by Expert
of Computer Vision.
Andrej Karpathy
Research Scientist at
OpenAI.

47. 47
Resources: Datasets
■ UCI Repository:
http://www.ics.uci.edu/~mlearn/MLRepository.html
■ UCI KDD Archive:
http://kdd.ics.uci.edu/summary.data.application.html
■ Statlib: http://lib.stat.cmu.edu/
■ Delve: http://www.cs.utoronto.ca/~delve/

48. 48
Resources: Journals
■ Journal of Machine Learning Research www.jmlr.org
■ IEEE Transactions on Neural Networks
■ IEEE Transactions on Pattern Analysis and Machine
Intelligence
■ Journal of the American Statistical Association
■ ...

49. 49
Resources: Conferences
■ International Conference on Machine Learning (ICML)
■ European Conference on Machine Learning (ECML)
■ Neural Information Processing Systems (NIPS)
■ Computational Learning
■ International Joint Conference on Artificial Intelligence (IJCAI)
■ ACM SIGKDD Conference on Knowledge Discovery and Data Mining
(KDD)
■ IEEE Int. Conf. on Data Mining (ICDM)

50. References
■ H. Bhaskar, D. Hoyle, and S. Singh (2006). Machine Learning: a
Brief Survey and Recommendations for Practitioners.
Computers in Biology and Medicine, 36(10), 1104-1125.
■ K. Church (1988). A Stochastic Parts Program and Noun Phrase
Parser for Unrestricted Texts. In Proc. ANLP-1988, 136-143.
■ S. Dumais, J. Platt, D. Heckerman and M. Sahami (1998).
Inductive Learning Algorithms and Representations for Text
Categorization. In Proc. CIKM-1998, 148-155.
■ K. Lee (1989). Automatic Speech Recognition: The
Development of the Sphinx System, Kluwer Academic
Publishers.
■ T. Mitchell (1997). Machine Learning, McGraw-Hill Publishers.
■ G. Tesauro (1995). Temporal Difference Learning and TD-
gammon. Communications of the ACM, 38(3), 58-68.
50

51. 51

52. 52