Professional Documents
Culture Documents
Introduction
Representations Matter
APTER 1. INTRODUCTION
x r
Composition
Output
CAR PERSON ANIMAL
(object identity)
Visible layer
(input pixels)
Figure 1.2: Illustration of a deep learning model. It is difficult for a computer to understand
Dr. CONVOLBO
Figure 1.2
the meaning of raw sensory input data, such as this image represented as a collection 3
Computational Graphs
CHAPTER 1. INTRODUCTION
Element Element
Set Set
+
+
⇥ ⇥ ⇥ Logistic
Regression
Logistic
Regression
w1 x1 w2 x2 w x
Representation learning
Machine learning
AI
Dr. CONVOLBO
Figure 1.4
Figure 1.4: A Venn diagram showing how deep learning is a kind of representation learning, 5
CHAPTER 1. INTRODUCTION
Learning Multiple
Components Output
Figure 1.5
Mapping from
Output Output
features
Additional
Mapping from Mapping from layers of more
Output
features features abstract
features
Hand- Hand-
Simple
designed designed Features
features
program features
Deep
Classic learning
Rule-based
machine
systems Representation
learning
Dr. CONVOLBO
learning 6
Organization of the Book
CHAPTER 1. INTRODUCTION
3. Probability and
2. Linear Algebra
Information Theory
6. Deep Feedforward
Networks
11. Practical
12. Applications
Methodology
18. Partition
19. Inference
Function
Dr. CONVOLBO 7
Figure 1.6: The high-level organization of the book. An arrow from one chapter to another
Historical Waves
0.000250
Frequency of Word or Phrase
cybernetics
0.000200
(connectionism + neural networks)
0.000150
0.000100
0.000050
0.000000
1940 1950 1960 1970 1980 1990 2000
Year
Dr. CONVOLBO
Figure 1.9
Figure 1.9: Example inputs from the MNIST dataset. The “NIST” stands for National 10
Connections per Neuron
104 Human
6 Cat
Connections per neuron
9 7
4
103 Mouse
2
10
5
8
102 Fruit fly
3
1
101
1950 1985 2000 2015
Year
Figure 1.10: Initially, the number of connections between neurons in artificial neura
networks was limited by hardware capabilities. Today, the number of connections between
neurons Figure
is mostly a design consideration.
Dr. CONVOLBO
1.10
Some artificial neural networks have nearly11a
Number of Neurons
Number of neurons (logarithmic scale)
1011 Human
1010
17 20
109 16 19 Octopus
108 14 18
107 11 Frog
106 8
105 3 Bee
Ant
104
103 Leech
13
102
101 1 2 12 15 Roundworm
6 9
100 5 10
10 1 4 7
10 2 Sponge
1950 1985 2000 2015 2056
Year
ure 1.11: Since the introduction of hidden units, artificial neural networks have doub
ize roughly every 2.4 years. Biological neural network sizes from Wikipedia (2015
1. Dr.Perceptron
CONVOLBO (Rosenblatt, 1958, 1962) Figure 1.11 12
Solving Object Recognition
0.30
ILSVRC classification error rate
0.25
0.20
0.15
0.10
0.05
0.00
2010 2011 2012 2013 2014 2015
Year
gure 1.12: Since deep networks reached the scale necessary to compete in the Ima
arge Scale Visual Recognition Challenge, they have consistently won the compe
ery year, and yielded lower and Figure
Dr. CONVOLBO lower error rates each time. Data from Russak
1.12 13