Chap 1.

Introduction

TS. Trần Vũ Hoàng

1
Welcome! – Facebook’s functions

2
Welcome! – Facebook’s functions

Recognize human’s name

Detect human’s face
Allow to tag their name

3
Welcome! – Advertisements

4
Welcome! – Mail’s function

5
Machine Learning and Its Applications

1 •What is Machine Learning?

2
•When should we apply Machine Learning?

3
•How can the Machine learn?

4 •Applications

6
What is Machine Learning?

What is the basic difference between human and computer?

7
What is Machine Learning?

Learn from experience Follow instructions

8
What is Machine Learning?

Learn from experience Learn from experience Follow instructions

9
What is Machine Learning?
data
Learn from experience Learn from experience Follow instructions

10
What is Machine Learning?

11
What is Machine Learning?

12
By journalist Michael Copeland, the fundamentals of deep learning
What is Machine Learning?

13
What is Machine Learning?
Traditional Programming

Machine Learning

14
What is Machine Learning?
• Machine Learning
• A field of AI
• The data analytics techniques
• Learn automatically from data to solve the specific tasks
• Ask computers to program themselves

15
What is Machine Learning?
• Arthur Samuel (1959). Machine Learning:
Field of study that gives computers the ability
to learn without being explicitly programmed.
• Tom Mitchell (1998) Well-posed Learning
Problem: A computer program is said to learn
from experience E with respect to some task
T and some performance measure P, if its
performance on T, as measured by P,
improves with experience E

16
 “A computer program is said to learn from experience E with respect
to some task T and some performance measure P, if its performance
on T, as measured by P, improves with experience E”

• Quiz: Suppose your email program watches which emails you do or

do not mark as spam, and based on that learns how to better filter
spam. What is the task T in this setting?
• Classifying emails as spam or not spam.
• Watching you label emails as spam or not spam.
• The number (or fraction) of emails correctly classified as spam/not spam.
• None of the above—this is not a machine learning problem.

17
 “A computer program is said to learn from experience E with respect
to some task T and some performance measure P, if its performance
on T, as measured by P, improves with experience E”

• Quiz: Suppose your email program watches which emails you do or

do not mark as spam, and based on that learns how to better filter
spam. What is the task T in this setting?
• Classifying emails as spam or not spam. T
• Watching you label emails as spam or not spam.
• The number (or fraction) of emails correctly classified as spam/not spam.
• None of the above—this is not a machine learning problem.

18
 “A computer program is said to learn from experience E with respect
to some task T and some performance measure P, if its performance
on T, as measured by P, improves with experience E”

• Quiz: Suppose your email program watches which emails you do or

do not mark as spam, and based on that learns how to better filter
spam. What is the task T in this setting?
• Classifying emails as spam or not spam. T
• Watching you label emails as spam or not spam. E
• The number (or fraction) of emails correctly classified as spam/not spam.
• None of the above—this is not a machine learning problem.

19
 “A computer program is said to learn from experience E with respect
to some task T and some performance measure P, if its performance
on T, as measured by P, improves with experience E”

• Quiz: Suppose your email program watches which emails you do or

do not mark as spam, and based on that learns how to better filter
spam. What is the task T in this setting?
• Classifying emails as spam or not spam. T
• Watching you label emails as spam or not spam. E
• The number (or fraction) of emails correctly classified as spam/not spam. P
• None of the above—this is not a machine learning problem.

20
Machine Learning and Its Applications

1 •What is Machine Learning?

2
•When should we apply Machine Learning?

3
•How can the Machine learn?

4 •Applications

21
When should we apply Machine Learning?

22
 When should we apply Machine Learning?
• Consider using Machine Learning when:
• Have a complex task
• Have a large amount of data
• Have a lot of variables
• But no existing formula or equation

23
Machine Learning and Its Applications

1 •What is Machine Learning?

2
•When should we apply Machine Learning?

3
•How can the Machine learn?

4 •Applications

24
How can the machine learn?

How can baby learn?

25
How can the machine learn?

26
How can the machine learn?

Supervised Learning 27
How can the machine learn?

Supervised Learning 28
How can the machine learn?

Supervised Learning Unsupervised Learning 29

How can the machine learn?

Supervised Learning Unsupervised Learning 30

How can the machine learn?

31
 Introduction
Supervised
Learning
Machine Learning
Supervised Learning

$70,000 ? $170,000
Supervised Learning
Price ($10,000)
20
17
15
12
10
7
5

200
Size of the house (m2)
Regression: Predict continuous valued output (price)
Supervised Learning

Classification: Discrete valued output (0 or 1)

You’re running a company, and you want to develop learning algorithms to address each
of two problems.

Problem 1: You have a large inventory of identical items. You want to predict how many
of these items will sell over the next 3 months.
Problem 2: You’d like software to examine individual customer accounts, and for each
account decide if it has been hacked/compromised.

Should you treat these as classification or as regression problems?

Treat both as classification problems.
Treat problem 1 as a classification problem, problem 2 as a regression problem.
Treat problem 1 as a regression problem, problem 2 as a classification problem.

Treat both as regression problems.

 Introduction
Unsupervised
Learning
Machine Learning
 Supervised Learning

x2

x1
 Unsupervised Learning

x2

x1
Classification

Clustering

40
41
Organize computing clusters Social network analysis

Image credit: NASA/JPL-Caltech/E. Churchwell (Univ. of Wisconsin, Madison)

Market segmentation Astronomical data analysis

Cocktail party problem

Speaker #1 Microphone #1

Speaker #2 Microphone #2
Of the following examples, which would you address using an
unsupervised learning algorithm? (Check all that apply.)

Given email labeled as spam/not spam, learn a spam filter.

Given a set of news articles found on the web, group them into
set of articles about the same story.
Given a database of customer data, automatically discover market
segments and group customers into different market segments.
Given a dataset of patients diagnosed as either having diabetes or
not, learn to classify new patients as having diabetes or not.
How can the machine learn?

Learn by trail and error 45

Machine Learning and Its Applications

1 •What is Machine Learning?

2
•When should we apply Machine Learning?

3
•How can the Machine learn?

4 •Applications

46
Applications
• Please list the applications that you think they are machine learning
problems and classify them into suitable categories:
• Supervised learning, classification
• Supervised learning, regression
• Unsupervised learning (Clustering)

47
Applications

48
Applications
• Computational finance, for credit scoring and algorithmic trading
• Image processing and computer vision, for face recognition, motion
detection, and object detection
• Computational biology, for tumor detection, drug discovery, and DNA
sequencing
• Energy production, for price and load forecasting
• Automotive, aerospace, and manufacturing, for predictive
maintenance
• Natural language processing, for voice recognition applications

49
Applications
üSelf-driving cars
üDeep Learning in Healthcare
üVoice Search & Voice-Activated Assistants
üAutomatically Adding Sounds To Silent Movies
üAutomatic Machine Translation
üAutomatic Text Generation
üAutomatic Handwriting Generation
üImage Recognition
üAutomatic Image Caption Generation

50
 Applications

üAutomatic Colorization

üAdvertising

üPredicting Earthquakes

üNeural Networks for Brain Cancer Detection

üNeural Networks in Finance

üEnergy Market Price Forecasting

Vartul Mittal, Top 15 Deep Learning applications that will rule the world in 2018 and beyond, [online] https://medium.com/@vratulmittal
51
 Machine Learning Projects for Beginners
üMachine Learning Gladiator
üPlay Money Ball
üPredict Stock Prices
üTeach a Neural Network to Read Handwriting
üInvestigate Enron
üWrite ML Algorithms from Scratch
üMine Social Media Sentiment
üImprove Health Care
Source: https://elitedatascience.com/machine-learning-projects-for-beginners
52
 Libraries for Deep Learning
• Tensorflow
• Keras
• Caffe
• Microsoft Cognitive Toolkit
• PyTorch
• …
 Useful tutorials
• https://kipalog.com/posts/Bat-dau-voi-Machine-Learning-thong-qua-
Tensorflow--Phan-I-2
• https://viblo.asia/p/deep-learning-qua-kho-dung-lo-da-co-keras-
LzD5dBqoZjY
• https://techwithtim.net/tutorials/machine-learning-
python/introduction/
References

• Pedro Domingos, Machine Learning, University of Washington, 2009

• Andrew Ng, Machine Learning, Stanford University [Online]

https://www.coursera.org

• Machine Learning, Mathworks [Online] https://www.mathworks.com

55
HomeWorks
• Install Python, Anaconda, IDE, Tensorflow

56