SlideShare a Scribd company logo

Introduction to machinel learning and deep learning

Download as PPTX, PDF
Ibrahim Amer, profile picture
Ibrahim Amer

The document provides an introduction to machine learning and deep learning. It defines machine learning as a field of artificial intelligence that uses statistical techniques to give computer systems the ability to "learn" with data, without being explicitly programmed. The document discusses different types of machine learning including supervised learning, unsupervised learning, and reinforcement learning. It also introduces deep learning and convolutional neural networks, describing how deep learning can be used to perform complex tasks like image recognition.

Education
Related topics:
Convolutional Neural Networks • Deep Learning
1 of 100
Downloaded 69 times
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
Introduction To Machine Learning and Deep Learning Ibrahim Amer TA at FCIS, Ain Shams University Ibrahim.ali.amer92@gmail.com 1
Agenda  What is the computer?  Introduction to machine learning  Applications on machine learning  Introduction to deep learning  Convolutional Neural Networks  How to build a computer to start deep learning  Deep learning Tools  Companies and deep learning 2
What is computer?  A computer is a machine that understands zeros and ones only.  The computer can perform complex mathematical and arithmetic operations very fast.  It performs this using a combination of circuits and logic gates. 3
Comment on the following  What is the result? 50 + 30 ? 4
Comment on the following  What is the result? 58945 + 78954? 5
Comment on the following  What is the result? 59545? 6
Comment on the following 7  What do you see?
Comment on the following 8  What do you see?
Comment on the following Can you write a program to compute the power ? 9
Comment on the following Can you write a program to recognize faces or recognize brands of cars? 10
Conclusion  Computers are very powerful and accurate in calculations, they can perform any arithmetic operation faster than any human being.  Humans are faster and more accurate than computers in recognition tasks.  Do you know any one who could compute 787452665 * 75767487 in just seconds?  Are the machines capable recognize and to reach human level accuracy in recognition ? 11
What is Machine Learning?  “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.” -- Tom Mitchell, Carnegie Mellon University 12
What is Machine Learning?  Machine learning is a discipline of AI  It is a series of techniques that are used to make the computer smarter .  ML solves problems that cannot be solved by numerical means alone.  Machine learning is empowering a lot of technologies used today.  Among the different types of ML tasks, a crucial distinction is drawn between supervised and unsupervised learning:  Supervised machine learning: The program is “trained” on a pre-defined set of “training examples”, which then facilitate its ability to reach an accurate conclusion when given new data.  Unsupervised machine learning: The program is given a bunch of data and must find patterns and relationships therein. 13
Supervised Machine Learning  In the majority of supervised learning applications, the ultimate goal is to develop a finely tuned predictor function ℎ(𝑥) (sometimes called the “hypothesis”)  Given input data 𝑥 about a certain domain (say, square footage of a house), it will accurately predict some interesting value h 𝑥  In practice, x almost always represents multiple data points. So, for example, a housing price predictor might take not only square-footage (x1) but also number of bedrooms (x2), number of bathrooms (x3), number of floors (x4), year built (x5), zip code (x6), and so forth. Determining which inputs to use is an important part of ML design. However, for the sake of explanation, it is easiest to assume a single input value is used. 14
 So let’s say our simple predictor has this form:  Where 𝜃0 and 𝜃1 are constants. Our goal is to find the perfect values 𝜃0 and 𝜃1 to make our predictor work as well as possible.  To make our program learn 𝜃0 and 𝜃1 we should update them according to error analysis which is called mean square error.  The weights are updated using gradient decent algorithm 15
16
17
Unsupervised Learning  Unsupervised learning typically is tasked with finding relationships within data.  There are no training examples used in this process.  Instead, the system is given a set data and tasked with finding patterns and correlations therein.  Examples: astronomical data analysis, social network analysis and cocktail party problem and news grouping 18
Applications on machine learning  Spam filtering: Classifying emails as spam or non-spam  Weather forecast: Machine learning is applied in weather forecasting software to improve the quality of the forecast.  Anti-virus: Machine learning is used in Anti-virus software's to improve detection of malicious software on computer devices.  Personal Assistants: Siri & Cortana  Classifying a tumor as benign tumor or malignant tumor. 19
Deep Learning  Deep Learning is a subfield of machine learning concerned with algorithms inspired by the structure and function of the brain called artificial neural networks.  Andrew Ng: “very large neural networks we can now have and … huge amounts of data that we have access to”  Andrew Ng: “for most flavors of the old generations of learning algorithms … performance will plateau. … deep learning … is the first class of algorithms … that is scalable. … performance just keeps getting better as you feed them more data” 20
Deep Learning 21
Deep Learning  Deep learning is set of techniques that in some cases could reach human accuracy in recognition !  The most common used and the most famous architecture is convolutional neural network.  There are some other techniques like:  LSTM (Long Short Term Memory)  Residual Neural Networks  Autoencoders  Generative Adversarial Networks.  We will focus on convolution neural networks. 22
A toy ConvNet: X’s and O’s X or OCNN A two-dimensional array of pixels
For example CNN X CNN O
Trickier cases CNN X CNN O
Deciding is hard ?
-1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 What computers see ?
-1 -1 -1 -1 -1 -1 -1 -1 -1 -1 X -1 -1 -1 -1 X X -1 -1 X X -1 -1 X X -1 -1 -1 -1 X 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 X -1 -1 -1 -1 X X -1 -1 X X -1 -1 X X -1 -1 -1 -1 X -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 What computers see
-1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Computers are literal x
ConvNets match pieces of the image = = =
1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 1 Features match pieces of the image
1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 1
1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 1
1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 1
1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 1
1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 1
1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
Filtering: The math behind the match 1. 2. 3. 4.
1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
1 1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
1 1 1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
1 1 1 1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
1 1 1 1 1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
1 1 1 1 1 1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
1 1 1 1 1 1 1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
1 1 1 1 1 1 1 1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
1 1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
1 1 -1 1 1 1 -1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 1 -1 1 1 1 -1 1 1 Filtering: The math behind the match 55 1 1 -1 1 1 1 -1 1 1
1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Convolution: Trying every possible match 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Convolution: Trying every possible match = 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
1 -1 -1 -1 1 -1 -1 -1 1 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 = 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 -1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 1 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.11 0.33 -0.77 1.00 -0.77 0.33 -0.11 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 = = -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
Convolution layer 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 1 -1 -1 -1 1 -1 -1 -1 1 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 -1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 1 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.11 0.33 -0.77 1.00 -0.77 0.33 -0.11 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
Convolution layer 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.11 0.33 -0.77 1.00 -0.77 0.33 -0.11 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
Pooling: Shrinking the image stack 1. 2. 3. 4.
1.00 Pooling 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
1.00 0.33 Pooling 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
1.00 0.33 0.55 Pooling 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
1.00 0.33 0.55 0.33 Pooling 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
1.00 0.33 0.55 0.33 0.33 Pooling 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
1.00 0.33 0.55 0.33 0.33 1.00 0.33 0.55 0.55 0.33 1.00 0.11 0.33 0.55 0.11 0.77 Pooling 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
1.00 0.33 0.55 0.33 0.33 1.00 0.33 0.55 0.55 0.33 1.00 0.11 0.33 0.55 0.11 0.77 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.11 0.33 -0.77 1.00 -0.77 0.33 -0.11 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 0.33 0.55 1.00 0.77 0.55 0.55 1.00 0.33 1.00 1.00 0.11 0.55 0.77 0.33 0.55 0.33 0.55 0.33 0.55 0.33 0.33 1.00 0.55 0.11 0.55 0.55 0.55 0.11 0.33 0.11 0.11 0.33
Pooling layer 1.00 0.33 0.55 0.33 0.33 1.00 0.33 0.55 0.55 0.33 1.00 0.11 0.33 0.55 0.11 0.77 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.11 0.33 -0.77 1.00 -0.77 0.33 -0.11 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 0.33 0.55 1.00 0.77 0.55 0.55 1.00 0.33 1.00 1.00 0.11 0.55 0.77 0.33 0.55 0.33 0.55 0.33 0.55 0.33 0.33 1.00 0.55 0.11 0.55 0.55 0.55 0.11 0.33 0.11 0.11 0.33
Normalization
Rectified Linear Units (ReLUs) 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 0.77
0.77 0 Rectified Linear Units (ReLUs) 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
0.77 0 0.11 0.33 0.55 0 0.33 Rectified Linear Units (ReLUs) 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
0.77 0 0.11 0.33 0.55 0 0.33 0 1.00 0 0.33 0 0.11 0 0.11 0 1.00 0 0.11 0 0.55 0.33 0.33 0 0.55 0 0.33 0.33 0.55 0 0.11 0 1.00 0 0.11 0 0.11 0 0.33 0 1.00 0 0.33 0 0.55 0.33 0.11 0 0.77 Rectified Linear Units (ReLUs) 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
ReLU layer 0.77 0 0.11 0.33 0.55 0 0.33 0 1.00 0 0.33 0 0.11 0 0.11 0 1.00 0 0.11 0 0.55 0.33 0.33 0 0.55 0 0.33 0.33 0.55 0 0.11 0 1.00 0 0.11 0 0.11 0 0.33 0 1.00 0 0.33 0 0.55 0.33 0.11 0 0.77 0.33 0 0.11 0 0.11 0 0.33 0 0.55 0 0.33 0 0.55 0 0.11 0 0.55 0 0.55 0 0.11 0 0.33 0 1.00 0 0.33 0 0.11 0 0.55 0 0.55 0 0.11 0 0.55 0 0.33 0 0.55 0 0.33 0 0.11 0 0.11 0 0.33 0.33 0 0.55 0.33 0.11 0 0.77 0 0.11 0 0.33 0 1.00 0 0.55 0 0.11 0 1.00 0 0.11 0.33 0.33 0 0.55 0 0.33 0.33 0.11 0 1.00 0 0.11 0 0.55 0 1.00 0 0.33 0 0.11 0 0.77 0 0.11 0.33 0.55 0 0.33 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.11 0.33 -0.77 1.00 -0.77 0.33 -0.11 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33
Layers get stacked -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1.00 0.33 0.55 0.33 0.33 1.00 0.33 0.55 0.55 0.33 1.00 0.11 0.33 0.55 0.11 0.77 0.33 0.55 1.00 0.77 0.55 0.55 1.00 0.33 1.00 1.00 0.11 0.55 0.77 0.33 0.55 0.33 0.55 0.33 0.55 0.33 0.33 1.00 0.55 0.11 0.55 0.55 0.55 0.11 0.33 0.11 0.11 0.33
Deep stacking -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1.00 0.55 0.55 1.00 0.55 1.00 1.00 0.55 1.00 0.55 0.55 0.55
Fully connected layer 1.00 0.55 0.55 1.00 0.55 1.00 1.00 0.55 1.00 0.55 0.55 0.55 1.00 0.55 0.55 1.00 1.00 0.55 0.55 0.55 0.55 1.00 1.00 0.55
Fully connected layer X O 1.00 0.55 0.55 1.00 1.00 0.55 0.55 0.55 0.55 1.00 1.00 0.55
Fully connected layer X O 0.55 1.00 1.00 0.55 0.55 0.55 0.55 0.55 1.00 0.55 0.55 1.00
Fully connected layer X O 0.9 0.65 0.45 0.87 0.96 0.73 0.23 0.63 0.44 0.89 0.94 0.53
Fully connected layer X O 0.9 0.65 0.45 0.87 0.96 0.73 0.23 0.63 0.44 0.89 0.94 0.53
Fully connected layer X O 0.9 0.65 0.45 0.87 0.96 0.73 0.23 0.63 0.44 0.89 0.94 0.53
Fully connected layer X O 0.9 0.65 0.45 0.87 0.96 0.73 0.23 0.63 0.44 0.89 0.94 0.53
Fully connected layer X O 0.9 0.65 0.45 0.87 0.96 0.73 0.23 0.63 0.44 0.89 0.94 0.53
Fully connected layer X O 0.9 0.65 0.45 0.87 0.96 0.73 0.23 0.63 0.44 0.89 0.94 0.53
Fully connected layer X O 0.9 0.65 0.45 0.87 0.96 0.73 0.23 0.63 0.44 0.89 0.94 0.53
Fully connected layer X O 0.9 0.65 0.45 0.87 0.96 0.73 0.23 0.63 0.44 0.89 0.94 0.53
Putting it all together -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 X O
How to build a computer to start deep learning ?  Deep learning techniques’ are resources hungry.  They require huge computing power to give you good performance.  Imagine we have 10 convolution layers of: 256 Conv Filters * Image(1000, 1000) 88
How to build a computer to start deep learning ?  This is pretty huge !.  If you ran this on a CPU it would take 1 day to process 60 000 training images on I7 processor!  What if the network is deeper than 20 layers and more complex than 256 conv filters / layer?  The network in this case may take days or even weeks to finish!! 89
So, what is the solution ?  In each layer in a CNN we are convolving the image with different filters.  What if we could convolve the image with different filters parallel in the same time?  The solution is to use GPU! 90
Comment on the following 91  How many cores inside a CPU ?  How many cores inside a GPU?
What are GPUs brands available out there?  There are two famous companies:  Unfortunately, AMD GPUs can’t be used for deep learning.  Nvidia GPUs only can be used  That’s because Nvidia provides tools and support for deep learning geeks.  There is a library called NVIDIA Cuda Toolkit 92
93
Nvidia GPUs  Almost all Nvidia’s GPUs will do the job.  You must check if it is CUDA enabled or not.  You can check online. Just search for CUDA supported GPUs.  Some of CUDA supported GPUs: TITAN X, Geforce GTX 1080, Geforce GTX 1070, Geforce GTX 1060, Geforce GTX 1050, Geforce GTX 980, Geforce GTX 970, Geforce GTX 960, Geforce GTX 9xx M Series, Geforce GTX 6xx Series, Geforce GTX 6xx M Series.  You can always use your CPU if you don’t have Nvidia GPU or if you don’t have GPU at all!. 94
Deep Learning Tools  After building your computer and configuring it for deep learning, you need the right tools and APIs to start coding  There are plenty of languages that can be used:  Matlab  Python  C++  C#  Java  And so many… 95
Deep Learning Tools  Deep learning frameworks:  NVIDIA CUDA programming APIs.  Tensorflow (Google’s library) for python.  Theano for python.  Torch (Facebook’s library) for Lua.  Caffe for python.  CNTK (Microsoft’s library) for python.  Keras to simplify coding for tensorflow and theano. 96
Deep Learning and Companies  Many Companies use deep learning on a daily basis.  Facebook: facebook auto tagger, videos and photos auto caption, post analyzer,  DeepMind (acquired by google): AlphaGo and many  Amazon 97
Questions ?! 98
References  Coursera’s Machine Learning course by Andrew Ng  CS231n: Convolutional Neural Networks for Visual Recognition  Setup a Deep Learning Environment on Windows (Theano & Keras with GPU Enabled)  ConvNets Visualization  How do Convolutional Neural Networks work?  Rana el Kaliouby, Co-founder, CEO at Affectiva  Hussein Mehanna 99
Thank you! 100

More Related Content

PPTX
Image Processing and Computer Vision
Silicon Mentor
 
PDF
MDEC Data Matters Series: machine learning and Deep Learning, A Primer
Poo Kuan Hoong
 
PDF
Introduction to Machine Learning and Deep Learning
Terry Taewoong Um
 
PPTX
Machine Learning, Deep Learning and Data Analysis Introduction
Te-Yen Liu
 
PPTX
Introduction to Machine Learning
Lior Rokach
 
PPTX
Machine Learning, Deep Learning how to use in civic tehnology
Kaz Furukawa
 
PPTX
Machine Learning and Deep Contemplation of Data
Joel Saltz
 
PDF
Lecture 1
Wael Sharba
 
Image Processing and Computer Vision
Silicon Mentor
 
MDEC Data Matters Series: machine learning and Deep Learning, A Primer
Poo Kuan Hoong
 
Introduction to Machine Learning and Deep Learning
Terry Taewoong Um
 
Machine Learning, Deep Learning and Data Analysis Introduction
Te-Yen Liu
 
Introduction to Machine Learning
Lior Rokach
 
Machine Learning, Deep Learning how to use in civic tehnology
Kaz Furukawa
 
Machine Learning and Deep Contemplation of Data
Joel Saltz
 
Lecture 1
Wael Sharba
 

Viewers also liked (14)

PDF
#6 PyData Warsaw: Deep learning for image segmentation
Matthew Opala
 
PDF
Video Analysis with Convolutional Neural Networks (Master Computer Vision Bar...
Universitat Politècnica de Catalunya
 
PPTX
Sf data mining_meetup
Adam Gibson
 
ODP
An Introduction to Computer Vision
guestd1b1b5
 
PPTX
COM2304: Introduction to Computer Vision & Image Processing
Hemantha Kulathilake
 
PPTX
Introduction to Machine Learning
James Ward
 
PDF
Deep Learning Cases: Text and Image Processing
Grigory Sapunov
 
PPTX
An introduction to Deep Learning
David Rostcheck
 
PPTX
An introduction to Machine Learning (and a little bit of Deep Learning)
Thomas da Silva Paula
 
PPTX
Artificial Intelligence, Machine Learning and Deep Learning
Sujit Pal
 
PPT
Computer Vision Basics
Suren Kumar
 
PDF
Deep Learning and Reinforcement Learning
Renārs Liepiņš
 
PDF
Transform your Business with AI, Deep Learning and Machine Learning
Sri Ambati
 
PDF
Data Science - Part XVII - Deep Learning & Image Processing
Derek Kane
 
#6 PyData Warsaw: Deep learning for image segmentation
Matthew Opala
 
Video Analysis with Convolutional Neural Networks (Master Computer Vision Bar...
Universitat Politècnica de Catalunya
 
Sf data mining_meetup
Adam Gibson
 
An Introduction to Computer Vision
guestd1b1b5
 
COM2304: Introduction to Computer Vision & Image Processing
Hemantha Kulathilake
 
Introduction to Machine Learning
James Ward
 
Deep Learning Cases: Text and Image Processing
Grigory Sapunov
 
An introduction to Deep Learning
David Rostcheck
 
An introduction to Machine Learning (and a little bit of Deep Learning)
Thomas da Silva Paula
 
Artificial Intelligence, Machine Learning and Deep Learning
Sujit Pal
 
Computer Vision Basics
Suren Kumar
 
Deep Learning and Reinforcement Learning
Renārs Liepiņš
 
Transform your Business with AI, Deep Learning and Machine Learning
Sri Ambati
 
Data Science - Part XVII - Deep Learning & Image Processing
Derek Kane
 
Ad

Similar to Introduction to machinel learning and deep learning (20)

PPTX
DL_Lecture_29_06_2020.pptx
AmlanChakrabarti3
 
PPTX
Deep Learning con CNTK by Pablo Doval
Plain Concepts
 
PPTX
Estado del Arte de la IA
Plain Concepts
 
PPTX
machine learning in the age of big data: new approaches and business applicat...
Armando Vieira
 
PDF
Machine learning and its parameter is discussed here
RevathiSundar4
 
PPT
Machine learning-in-details-with-out-python-code
Osama Ghandour Geris
 
PDF
Introduction to machine learning NYU.pdf
fankerui92
 
PDF
Deep learning_ adaptive computation and machine learning ( PDFDrive ).pdf
Sanjay Singh Nirwan
 
PPTX
ML Lec 1 (1).pptx
MuhammadTalha278665
 
DOC
learningIntro.doc
butest
 
DOC
learningIntro.doc
butest
 
PPT
Eick/Alpaydin Introduction
butest
 
PDF
Machine Learning - Implementation with Python - 1
University College of Engineering Kakinada, JNTUK - Kakinada, India
 
PPTX
machine _learning_introductionand python.pptx
ChandrakalaV15
 
PPTX
An introduction to machine learning
Avinash Kumar
 
PPTX
Machine learning
Vikas Sinha
 
PDF
Geometric Processing of Data in Neural Networks
Lorenzo Cassani
 
PPTX
Machine Learning.pptx
chadhar227
 
PDF
Prepare your data for machine learning
Ivo Andreev
 
PPTX
Machine learning
eonx_32
 
DL_Lecture_29_06_2020.pptx
AmlanChakrabarti3
 
Deep Learning con CNTK by Pablo Doval
Plain Concepts
 
Estado del Arte de la IA
Plain Concepts
 
machine learning in the age of big data: new approaches and business applicat...
Armando Vieira
 
Machine learning and its parameter is discussed here
RevathiSundar4
 
Machine learning-in-details-with-out-python-code
Osama Ghandour Geris
 
Introduction to machine learning NYU.pdf
fankerui92
 
Deep learning_ adaptive computation and machine learning ( PDFDrive ).pdf
Sanjay Singh Nirwan
 
ML Lec 1 (1).pptx
MuhammadTalha278665
 
learningIntro.doc
butest
 
learningIntro.doc
butest
 
Eick/Alpaydin Introduction
butest
 
Machine Learning - Implementation with Python - 1
University College of Engineering Kakinada, JNTUK - Kakinada, India
 
machine _learning_introductionand python.pptx
ChandrakalaV15
 
An introduction to machine learning
Avinash Kumar
 
Machine learning
Vikas Sinha
 
Geometric Processing of Data in Neural Networks
Lorenzo Cassani
 
Machine Learning.pptx
chadhar227
 
Prepare your data for machine learning
Ivo Andreev
 
Machine learning
eonx_32
 
Ad

Recently uploaded (20)

PDF
Insiders guide to clinical Medicine.pdf
AbhishekPatil315128
 
PPTX
Renaissance Architecture: A Journey from Faith to Humanism
DrMonicaSharma
 
PDF
O7-L3 Supply Chain Operations - ICLT Program
labyankof
 
PDF
3rd Neelam Sanjeevareddy Memorial Lecture.pdf
Academy of Grassroots Studies and Research of India (AGRASRI)
 
PPTX
school management -TNTEU- B.Ed., Semester II Unit 1.pptx
NihumathunnisaH
 
PDF
Classroom Observation Tools for Teachers
Nicaramirez
 
PPTX
Final Presentation General Medicine 03-08-2024.pptx
ZaheerAhmad228692
 
PPTX
Cell Structure & Organelles in detailed.
DHANASHREESIVAKUMAR
 
PDF
Microbial disease of the cardiovascular and lymphatic systems
KeyaSharma
 
PPTX
GDM (1) (1).pptx small presentation for students
shrawanbpkihs
 
PPTX
Cell Types and Its function , kingdom of life
ClaireMangundayao1
 
PPTX
Lesson notes of climatology university.
sgladness67
 
PDF
Saundersa Comprehensive Review for the NCLEX-RN Examination.pdf
sreejithcareers
 
PDF
Supply Chain Operations Speaking Notes -ICLT Program
labyankof
 
PDF
Basic Mud Logging Guide for educational purpose
wdandworks
 
PDF
grade 11-chemistry_fetena_net_5883.pdf teacher guide for all student
banteamlakmebratu738
 
PDF
TR - Agricultural Crops Production NC III.pdf
avgilmegino3
 
PDF
Abdominal Access Techniques with Prof. Dr. R K Mishra
mohitsuren827
 
PDF
Sports Quiz easy sports quiz sports quiz
nikunj2757
 
PPTX
Institutional Correction lecture only . . .
limvtheghins
 
Insiders guide to clinical Medicine.pdf
AbhishekPatil315128
 
Renaissance Architecture: A Journey from Faith to Humanism
DrMonicaSharma
 
O7-L3 Supply Chain Operations - ICLT Program
labyankof
 
3rd Neelam Sanjeevareddy Memorial Lecture.pdf
Academy of Grassroots Studies and Research of India (AGRASRI)
 
school management -TNTEU- B.Ed., Semester II Unit 1.pptx
NihumathunnisaH
 
Classroom Observation Tools for Teachers
Nicaramirez
 
Final Presentation General Medicine 03-08-2024.pptx
ZaheerAhmad228692
 
Cell Structure & Organelles in detailed.
DHANASHREESIVAKUMAR
 
Microbial disease of the cardiovascular and lymphatic systems
KeyaSharma
 
GDM (1) (1).pptx small presentation for students
shrawanbpkihs
 
Cell Types and Its function , kingdom of life
ClaireMangundayao1
 
Lesson notes of climatology university.
sgladness67
 
Saundersa Comprehensive Review for the NCLEX-RN Examination.pdf
sreejithcareers
 
Supply Chain Operations Speaking Notes -ICLT Program
labyankof
 
Basic Mud Logging Guide for educational purpose
wdandworks
 
grade 11-chemistry_fetena_net_5883.pdf teacher guide for all student
banteamlakmebratu738
 
TR - Agricultural Crops Production NC III.pdf
avgilmegino3
 
Abdominal Access Techniques with Prof. Dr. R K Mishra
mohitsuren827
 
Sports Quiz easy sports quiz sports quiz
nikunj2757
 
Institutional Correction lecture only . . .
limvtheghins
 

Introduction to machinel learning and deep learning

  • 1. Introduction To Machine Learning and Deep Learning Ibrahim Amer TA at FCIS, Ain Shams University Ibrahim.ali.amer92@gmail.com 1
  • 2. Agenda  What is the computer?  Introduction to machine learning  Applications on machine learning  Introduction to deep learning  Convolutional Neural Networks  How to build a computer to start deep learning  Deep learning Tools  Companies and deep learning 2
  • 3. What is computer?  A computer is a machine that understands zeros and ones only.  The computer can perform complex mathematical and arithmetic operations very fast.  It performs this using a combination of circuits and logic gates. 3
  • 4. Comment on the following  What is the result? 50 + 30 ? 4
  • 5. Comment on the following  What is the result? 58945 + 78954? 5
  • 6. Comment on the following  What is the result? 59545? 6
  • 7. Comment on the following 7  What do you see?
  • 8. Comment on the following 8  What do you see?
  • 9. Comment on the following Can you write a program to compute the power ? 9
  • 10. Comment on the following Can you write a program to recognize faces or recognize brands of cars? 10
  • 11. Conclusion  Computers are very powerful and accurate in calculations, they can perform any arithmetic operation faster than any human being.  Humans are faster and more accurate than computers in recognition tasks.  Do you know any one who could compute 787452665 * 75767487 in just seconds?  Are the machines capable recognize and to reach human level accuracy in recognition ? 11
  • 12. What is Machine Learning?  “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.” -- Tom Mitchell, Carnegie Mellon University 12
  • 13. What is Machine Learning?  Machine learning is a discipline of AI  It is a series of techniques that are used to make the computer smarter .  ML solves problems that cannot be solved by numerical means alone.  Machine learning is empowering a lot of technologies used today.  Among the different types of ML tasks, a crucial distinction is drawn between supervised and unsupervised learning:  Supervised machine learning: The program is “trained” on a pre-defined set of “training examples”, which then facilitate its ability to reach an accurate conclusion when given new data.  Unsupervised machine learning: The program is given a bunch of data and must find patterns and relationships therein. 13
  • 14. Supervised Machine Learning  In the majority of supervised learning applications, the ultimate goal is to develop a finely tuned predictor function ℎ(𝑥) (sometimes called the “hypothesis”)  Given input data 𝑥 about a certain domain (say, square footage of a house), it will accurately predict some interesting value h 𝑥  In practice, x almost always represents multiple data points. So, for example, a housing price predictor might take not only square-footage (x1) but also number of bedrooms (x2), number of bathrooms (x3), number of floors (x4), year built (x5), zip code (x6), and so forth. Determining which inputs to use is an important part of ML design. However, for the sake of explanation, it is easiest to assume a single input value is used. 14
  • 15.  So let’s say our simple predictor has this form:  Where 𝜃0 and 𝜃1 are constants. Our goal is to find the perfect values 𝜃0 and 𝜃1 to make our predictor work as well as possible.  To make our program learn 𝜃0 and 𝜃1 we should update them according to error analysis which is called mean square error.  The weights are updated using gradient decent algorithm 15
  • 16. 16
  • 17. 17
  • 18. Unsupervised Learning  Unsupervised learning typically is tasked with finding relationships within data.  There are no training examples used in this process.  Instead, the system is given a set data and tasked with finding patterns and correlations therein.  Examples: astronomical data analysis, social network analysis and cocktail party problem and news grouping 18
  • 19. Applications on machine learning  Spam filtering: Classifying emails as spam or non-spam  Weather forecast: Machine learning is applied in weather forecasting software to improve the quality of the forecast.  Anti-virus: Machine learning is used in Anti-virus software's to improve detection of malicious software on computer devices.  Personal Assistants: Siri & Cortana  Classifying a tumor as benign tumor or malignant tumor. 19
  • 20. Deep Learning  Deep Learning is a subfield of machine learning concerned with algorithms inspired by the structure and function of the brain called artificial neural networks.  Andrew Ng: “very large neural networks we can now have and … huge amounts of data that we have access to”  Andrew Ng: “for most flavors of the old generations of learning algorithms … performance will plateau. … deep learning … is the first class of algorithms … that is scalable. … performance just keeps getting better as you feed them more data” 20
  • 22. Deep Learning  Deep learning is set of techniques that in some cases could reach human accuracy in recognition !  The most common used and the most famous architecture is convolutional neural network.  There are some other techniques like:  LSTM (Long Short Term Memory)  Residual Neural Networks  Autoencoders  Generative Adversarial Networks.  We will focus on convolution neural networks. 22
  • 23. A toy ConvNet: X’s and O’s X or OCNN A two-dimensional array of pixels
  • 27. -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 What computers see ?
  • 28. -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 X -1 -1 -1 -1 X X -1 -1 X X -1 -1 X X -1 -1 -1 -1 X 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 X -1 -1 -1 -1 X X -1 -1 X X -1 -1 X X -1 -1 -1 -1 X -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 What computers see
  • 29. -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Computers are literal x
  • 30. ConvNets match pieces of the image = = =
  • 31. 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 1 Features match pieces of the image
  • 32. 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 1
  • 33. 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 1
  • 34. 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 1
  • 35. 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 1
  • 36. 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 1
  • 37. 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
  • 38. Filtering: The math behind the match 1. 2. 3. 4.
  • 39. 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
  • 40. 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
  • 41. 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
  • 42. 1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
  • 43. 1 1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
  • 44. 1 1 1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
  • 45. 1 1 1 1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
  • 46. 1 1 1 1 1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
  • 47. 1 1 1 1 1 1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
  • 48. 1 1 1 1 1 1 1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
  • 49. 1 1 1 1 1 1 1 1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
  • 50. 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
  • 51. 1 1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
  • 52. 1 1 -1 1 1 1 -1 1 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Filtering: The math behind the match
  • 53. 1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 1 -1 1 1 1 -1 1 1 Filtering: The math behind the match 55 1 1 -1 1 1 1 -1 1 1
  • 54. 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Convolution: Trying every possible match 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
  • 55. 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 Convolution: Trying every possible match = 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
  • 56. 1 -1 -1 -1 1 -1 -1 -1 1 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 = 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 -1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 1 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.11 0.33 -0.77 1.00 -0.77 0.33 -0.11 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 = = -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
  • 57. Convolution layer 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 1 -1 -1 -1 1 -1 -1 -1 1 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 -1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 1 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.11 0.33 -0.77 1.00 -0.77 0.33 -0.11 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
  • 58. Convolution layer 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.11 0.33 -0.77 1.00 -0.77 0.33 -0.11 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
  • 59. Pooling: Shrinking the image stack 1. 2. 3. 4.
  • 60. 1.00 Pooling 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
  • 61. 1.00 0.33 Pooling 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
  • 62. 1.00 0.33 0.55 Pooling 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
  • 63. 1.00 0.33 0.55 0.33 Pooling 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
  • 64. 1.00 0.33 0.55 0.33 0.33 Pooling 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
  • 65. 1.00 0.33 0.55 0.33 0.33 1.00 0.33 0.55 0.55 0.33 1.00 0.11 0.33 0.55 0.11 0.77 Pooling 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
  • 66. 1.00 0.33 0.55 0.33 0.33 1.00 0.33 0.55 0.55 0.33 1.00 0.11 0.33 0.55 0.11 0.77 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.11 0.33 -0.77 1.00 -0.77 0.33 -0.11 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 0.33 0.55 1.00 0.77 0.55 0.55 1.00 0.33 1.00 1.00 0.11 0.55 0.77 0.33 0.55 0.33 0.55 0.33 0.55 0.33 0.33 1.00 0.55 0.11 0.55 0.55 0.55 0.11 0.33 0.11 0.11 0.33
  • 67. Pooling layer 1.00 0.33 0.55 0.33 0.33 1.00 0.33 0.55 0.55 0.33 1.00 0.11 0.33 0.55 0.11 0.77 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.11 0.33 -0.77 1.00 -0.77 0.33 -0.11 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 0.33 0.55 1.00 0.77 0.55 0.55 1.00 0.33 1.00 1.00 0.11 0.55 0.77 0.33 0.55 0.33 0.55 0.33 0.55 0.33 0.33 1.00 0.55 0.11 0.55 0.55 0.55 0.11 0.33 0.11 0.11 0.33
  • 69. Rectified Linear Units (ReLUs) 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 0.77
  • 70. 0.77 0 Rectified Linear Units (ReLUs) 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
  • 71. 0.77 0 0.11 0.33 0.55 0 0.33 Rectified Linear Units (ReLUs) 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
  • 72. 0.77 0 0.11 0.33 0.55 0 0.33 0 1.00 0 0.33 0 0.11 0 0.11 0 1.00 0 0.11 0 0.55 0.33 0.33 0 0.55 0 0.33 0.33 0.55 0 0.11 0 1.00 0 0.11 0 0.11 0 0.33 0 1.00 0 0.33 0 0.55 0.33 0.11 0 0.77 Rectified Linear Units (ReLUs) 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77
  • 73. ReLU layer 0.77 0 0.11 0.33 0.55 0 0.33 0 1.00 0 0.33 0 0.11 0 0.11 0 1.00 0 0.11 0 0.55 0.33 0.33 0 0.55 0 0.33 0.33 0.55 0 0.11 0 1.00 0 0.11 0 0.11 0 0.33 0 1.00 0 0.33 0 0.55 0.33 0.11 0 0.77 0.33 0 0.11 0 0.11 0 0.33 0 0.55 0 0.33 0 0.55 0 0.11 0 0.55 0 0.55 0 0.11 0 0.33 0 1.00 0 0.33 0 0.11 0 0.55 0 0.55 0 0.11 0 0.55 0 0.33 0 0.55 0 0.33 0 0.11 0 0.11 0 0.33 0.33 0 0.55 0.33 0.11 0 0.77 0 0.11 0 0.33 0 1.00 0 0.55 0 0.11 0 1.00 0 0.11 0.33 0.33 0 0.55 0 0.33 0.33 0.11 0 1.00 0 0.11 0 0.55 0 1.00 0 0.33 0 0.11 0 0.77 0 0.11 0.33 0.55 0 0.33 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 0.77 -0.11 0.11 0.33 0.55 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.11 -0.11 0.11 -0.11 1.00 -0.33 0.11 -0.11 0.55 0.33 0.33 -0.33 0.55 -0.33 0.33 0.33 0.55 -0.11 0.11 -0.33 1.00 -0.11 0.11 -0.11 0.11 -0.11 0.33 -0.11 1.00 -0.11 0.33 -0.11 0.55 0.33 0.11 -0.11 0.77 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.11 0.33 -0.77 1.00 -0.77 0.33 -0.11 0.11 -0.55 0.55 -0.77 0.55 -0.55 0.11 -0.55 0.55 -0.55 0.33 -0.55 0.55 -0.55 0.33 -0.55 0.11 -0.11 0.11 -0.55 0.33
  • 74. Layers get stacked -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1.00 0.33 0.55 0.33 0.33 1.00 0.33 0.55 0.55 0.33 1.00 0.11 0.33 0.55 0.11 0.77 0.33 0.55 1.00 0.77 0.55 0.55 1.00 0.33 1.00 1.00 0.11 0.55 0.77 0.33 0.55 0.33 0.55 0.33 0.55 0.33 0.33 1.00 0.55 0.11 0.55 0.55 0.55 0.11 0.33 0.11 0.11 0.33
  • 75. Deep stacking -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1.00 0.55 0.55 1.00 0.55 1.00 1.00 0.55 1.00 0.55 0.55 0.55
  • 76. Fully connected layer 1.00 0.55 0.55 1.00 0.55 1.00 1.00 0.55 1.00 0.55 0.55 0.55 1.00 0.55 0.55 1.00 1.00 0.55 0.55 0.55 0.55 1.00 1.00 0.55
  • 87. Putting it all together -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 X O
  • 88. How to build a computer to start deep learning ?  Deep learning techniques’ are resources hungry.  They require huge computing power to give you good performance.  Imagine we have 10 convolution layers of: 256 Conv Filters * Image(1000, 1000) 88
  • 89. How to build a computer to start deep learning ?  This is pretty huge !.  If you ran this on a CPU it would take 1 day to process 60 000 training images on I7 processor!  What if the network is deeper than 20 layers and more complex than 256 conv filters / layer?  The network in this case may take days or even weeks to finish!! 89
  • 90. So, what is the solution ?  In each layer in a CNN we are convolving the image with different filters.  What if we could convolve the image with different filters parallel in the same time?  The solution is to use GPU! 90
  • 91. Comment on the following 91  How many cores inside a CPU ?  How many cores inside a GPU?
  • 92. What are GPUs brands available out there?  There are two famous companies:  Unfortunately, AMD GPUs can’t be used for deep learning.  Nvidia GPUs only can be used  That’s because Nvidia provides tools and support for deep learning geeks.  There is a library called NVIDIA Cuda Toolkit 92
  • 93. 93
  • 94. Nvidia GPUs  Almost all Nvidia’s GPUs will do the job.  You must check if it is CUDA enabled or not.  You can check online. Just search for CUDA supported GPUs.  Some of CUDA supported GPUs: TITAN X, Geforce GTX 1080, Geforce GTX 1070, Geforce GTX 1060, Geforce GTX 1050, Geforce GTX 980, Geforce GTX 970, Geforce GTX 960, Geforce GTX 9xx M Series, Geforce GTX 6xx Series, Geforce GTX 6xx M Series.  You can always use your CPU if you don’t have Nvidia GPU or if you don’t have GPU at all!. 94
  • 95. Deep Learning Tools  After building your computer and configuring it for deep learning, you need the right tools and APIs to start coding  There are plenty of languages that can be used:  Matlab  Python  C++  C#  Java  And so many… 95
  • 96. Deep Learning Tools  Deep learning frameworks:  NVIDIA CUDA programming APIs.  Tensorflow (Google’s library) for python.  Theano for python.  Torch (Facebook’s library) for Lua.  Caffe for python.  CNTK (Microsoft’s library) for python.  Keras to simplify coding for tensorflow and theano. 96
  • 97. Deep Learning and Companies  Many Companies use deep learning on a daily basis.  Facebook: facebook auto tagger, videos and photos auto caption, post analyzer,  DeepMind (acquired by google): AlphaGo and many  Amazon 97
  • 99. References  Coursera’s Machine Learning course by Andrew Ng  CS231n: Convolutional Neural Networks for Visual Recognition  Setup a Deep Learning Environment on Windows (Theano & Keras with GPU Enabled)  ConvNets Visualization  How do Convolutional Neural Networks work?  Rana el Kaliouby, Co-founder, CEO at Affectiva  Hussein Mehanna 99

Editor's Notes

  • #6: 1144
  • #21: GIVE A BRIEF TALK ABOUT NEURAL NETWORKS