Introduction to AI and Machine

Learning

1
 Learning Goals
In this section, we will cover:

- Definitions
- Machine Learning Terminology
- Types of Machine Learning
- Deep Learning
- History of AI

2
 Artificial Intelligence & Machine Learning
In this overview, we will discuss:
- Define "Artificial Intelligence” (AI), "Machine
Learning” (ML), and “Deep Learning” (DL).
- Explain how DL helps solve classical ML limitations.
- Explain key historical developments, and the “hype-
Al winter cycle.”
- Differentiate modern Al from prior Al.
- Relate sample applications of AI.
Al robot

3
Al Breakthroughs

4
 Al is the New Electricity

“About 100 years ago, electricity

transformed every major industry. Al
has advanced to the point where it
has the power to transform … every
major sector in coming years.”
-Andrew Ng,
Stanford University

5
 Definitions

- Artificial Intelligence (AI)

- Machine Learning (ML)
- Deep Learning (DL)

6
 Artificial Intelligence
“A branch of computer science dealing with the simulation of
intelligent behavior in computers.” - Merriam-Webster
“Colloquially, the term ‘artificial intelligence’ is applied when a
machine mimics ‘cognitive’ functions that humans associate
with other human minds, such as ‘learning’ and ‘problem
solving’.” - Wikipedia

7
Machine Learning

“The study and construction of

programs that are not
explicitly programmed, but
learn patterns as they are
exposed to more data over
time.”

8
 Machine Learning

These programs learn from repeatedly seeing data,

rather than being explicitly programmed by humans.

9
 Machine Learning Terminology
In this example, we learn to classify flower species from a set of measurement features.

sepal sepal width petal length petal width species

length
Features
6.7 3.0 5.2 2.3 virginica
(attributes of the data)
6.4 2.8 5.6 2.1 virginica

4.6 3.4 1.4 0.3 setosa

Target
6.9 3.1 4.9 1.5 versicolor
(column to be predicted)
4.4 2.9 1.4 0.2 setosa

4.8 3.0 1.4 0.1 setosa

5.9 3.0 5.1 1.8 virginica

5.4 3.9 1.3 0.4 setosa

4.9 3.0 1.4 0.2 setosa 10

Two Main Types of Machine Learning

11
 Machine Learning Example
- Suppose you wanted to identify
fraudulent credit card transactions.
- You could define features to be:
● Transaction time
● Transaction amount
● Transaction location
● Category of purchase

- The algorithm could learn what feature Credit card transactions

combinations suggest unusual activity.

12
 Machine Learning Limitations

- Suppose you wanted to determine if an

image is of a cat or a dog.
- What features would you use?
- This is where Deep Learning can come in.

13
Deep Learning
“Machine learning that involves
using very complicated models
called “deep neural networks”.

Models determine best

representation of original data;
in classic machine learning,
humans must do this.

14
Deep Learning Example

15
 History of AI
Al has experienced several hype cycles, where it has
oscillated between periods of excitement and disappointment.

16
Deep Learning Example

17
 History of AI
Al has experienced several hype cycles, where it has
oscillated between periods of excitement and disappointment.

18
 History of AI
Al has experienced several hype cycles, where it has
oscillated between periods of excitement and disappointment.

19
 History of AI
Al has experienced several hype cycles, where it has
oscillated between periods of excitement and disappointment.

20
 History of AI
Al has experienced several hype cycles, where it has
oscillated between periods of excitement and disappointment.

21
 History of AI
Al has experienced several hype cycles, where it has
oscillated between periods of excitement and disappointment.

22
 History of AI
Al has experienced several hype cycles, where it has
oscillated between periods of excitement and disappointment.

23
 1950s: Early AI
● 1950: Alan Turing developed the Turing
test, to test a machine's ability to exhibit
intelligent behavior.
● 1956: Artificial Intelligence was accepted
as a field at the Dartmouth Conference.
● 1957: Frank Rosenblatt invented the
perceptron algorithm. This was the
precursor to modern neural networks.

● 1959: Arthur Samuel published an

algorithm for a checkers program using
machine learning. 24
 The First “Al Winter"
● 1966: ALPAC committee evaluated Al
techniques for machine translation and
determined there was little yield from the
investment.
● 1969: Marvin Minsky published a book on the
limitations of the Perceptron algorithm which
slowed research in neural networks.
● 1973: The Lighthill report highlights Al's failure
to live up to promises.
The two reports led to cuts in government funding John R. Pierce, head of ALPAC
for Al research, leading to the first “Al Winter". 25
 1980's Al Boom
Expert Systems - systems with programmed rules
designed to mimic human experts.
- Ran on mainframe computers with specialized
programming languages (e.g. LISP).
- Were the first widely-used Al technology, with
two-thirds of "Fortune 500" companies using
them at their peak.
- 1986: The “Backpropagation" algorithm is able
to train multi-layer perceptrons, leading to new
successes and interest in neural network
research.
26
 Another Al Winter (1980s –1990s)
Expert systems' progress on solving business
problems slowed.
- Expert systems began to be melded into
software suites of general business
applications (e.g. SAP®, Oracle®) that could
run on PCs instead of mainframes.
- Neural networks didn't scale to large problems.
Software companies
- Interest in Al in business declined.

27
 1990s - 2000s: Machine Learning
Al solutions had successes in speech recognition,
medical diagnosis, robotics, and many other areas.
- Al algorithms were integrated into larger
systems and became useful throughout
industry.
- The Deep Blue chess system beat world chess
champion Garry Kasparov.
- Google's search engine launched using artificial
intelligence technology.

28
 2006: Rise of Deep Learning
● 2006: Geoffrey Hinton publishes a paper on
unsupervised pre-training that allowed deeper neural
networks to be trained.
- Neural networks are rebranded as deep learning.
● 2009: The ImageNet database of human-tagged
images is presented at the CVPR conference.
● 2010: Algorithms compete on several visual
recognition tasks at the first ImageNet competition.

29
 Deep Learning Breakthroughs (2012 – ?)
● 2012: Deep learning beats previous
benchmark on the imageNet competition.
● 2013: Deep learning is used to understand
"conceptual meaning" of words.
● 2014: Similar breakthroughs appeared in
language translation.

These have led to advancements in Web

Search, Document Search, Document
Summarization, and Machine Translation.
30
 Deep Learning Breakthroughs (2012 – ?)
● 2014: Stanford team creates computer vision
algorithm that can describe photos.
● 2015: Deep learning platform TensorFlow is
developed.
● 2016: DeepMind’s AlphaGo, developed by Aja
Huang, beats Go master Lee Se-dol.
● 2018: Waymo launches commercial self-
driving-car service in suburbs of Phoenix.
● 2019: IBM Project Debater is able to have a
full debate with rebuttal with champion human
debater. 31
 Summary
Introduction to Artificial Intelligence and Machine Learning
● Artificial Intelligence is a branch of computer science dealing with the
simulation of intelligent behavior in computers. Machines mimic
cognitive functions such as learning and problem solving.
● Machine learning is the study of programs that are not explicitly
programmed, but instead these algorithms learn patterns from data.
● Deep learning is a subset of machine learning in which multilayered
neural networks learn from vast amounts of data.

32
 Summary
History of AI
● AI has experienced cycles of AI winters and AI booms.
● AI solutions include speech recognition, computer vision, assisted
medical diagnosis, robotics, and others.

33
 Learning Recap
In this section, we learned:

- Definitions
- Machine Learning Terminology
- Types of Machine Learning
- Deep Learning
- History of AI

34