What is Intelligence?
■ In the popular sense, intelligence is
defined as the ability to learn and apply
knowledge to manipulate your
environment, as well as the ability to
reason and have abstract thought.
■ Other definitions include:
– Adaptability to a new environment
– The ability to evaluate and judge
– The ability to comprehend complex ideas
– The capacity for original and productive
thought
– The ability to learn quickly and learn from 1
What is Artificial Intelligence?
■ Artificial Intelligence is the study of how to make
computers do things at which, at the moment,
people are better.
■ Embedding human intelligence in to the machine so
that they can think like humans, behave like humans
and make decisions like humans.
2
Tasks: Machines better than humans
1. Speed
■ Computer can perform any calculation in
split of a second.
■ Human consumes more time to do the
same task.
3
Tasks: Machines better than humans
2. Accuracy
■ Computer are very much accurate unless any software or
hardware failure occurs.
■ Computer accurately performs what it is told to do.
■ Human is prone to errors.
4
Tasks: Machines better than humans
3. Delegacy
■ Computer can perform specific task repetitively infinite
number of times with uniform accuracy.
■ Human gets bored with repetitive tasks.
5
Tasks: Machines better than humans
4. Run for a long period of time
■ Computer can run for a very long period of
time (once started).
■ It does not need any rest.
■ Human needs some rest and energy to do a
task.
6
Tasks: Machines better than humans
5. Multitasking
■ Computer can perform several different tasks at a time at
very high speed and tremendous accuracy.
■ Human can not concentrate mind on multiple tasks.
7
Tasks: Humans better than machines
1. Flexibility
■ Computer are designed to do some specific tasks.
■ Computer can play music or print the document but it cannot
make a cup of tea foryou.
■ Humans are flexible and can perform wide variety of tasks.
8
Tasks: Humans better than machines
2. Thinking (Common Sense)
■ Computer always need details of operation to perform
processing on data (It does not think).
■ Human can make sense of operation and does not need details
every time. (Common Sense)
9
Tasks: Humans better than machines
3. Does not need continuously energy
■ The computer needs continuous electric supply to operate.
■ While a human also needs some sort of energy in the form of
food but not all the time.
10
Tasks: Humans better than machines
4. Infrequent Loss of data
■ The data store in a computer can be corrupted, deleted
accidently or infected by virus.
■ Human may also lose the memory when some natural disaster
attacks the memory.
(But It happens rarely)
11
Rectangles ------ Voila-Jones
method
What is Intelligence?
■ In the popular sense, intelligence is
defined as the ability to learn and
apply knowledge to manipulate your
environment, as well as the ability to
reason and have abstract thought.
■ Other definitions include:
– Adaptability to a new environment
– The ability to evaluate and judge
– The ability to comprehend complex ideas
– The capacity for original and productive
thought
– The ability to learn quickly and learn 1
What is Artificial Intelligence?
■ Artificial Intelligence is the study of how to
make computers do things at which, at the
moment, people are better.
■ Embedding human intelligence in to the
machine so that they can think like humans,
behave like humans and make decisions like
humans.
2
Machine Learning
Machine Learning definition
• Arthur Samuel (1959). Machine Learning:
Field of study that gives computers the ability
to learn without being explicitly programmed.
Types of Machine learning algorithms:
- Supervised learning
- Unsupervised learning
- Reinforcement learning
Supervised Learning
• Supervised learning is commonly used
in real world applications, such as face
and speech recognition
• Labeled data **
• There are two types:
• Classification.
• Regression
(c) Alexander Ihler
Unsupervised Learning
• Unsupervised learning is used to detect
anomalies, outliers, such as fraud or
defective equipment
• No labeled data
(c) Alexander Ihler
Reinforcement Learning
• Here learning data gives feedback so
that the system adjusts to dynamic
conditions in order to achieve a certain
objective.
(c) Alexander Ihler
Supervised Learning
Housing price prediction
400
300
Price ($)
in 1000’s 200
100
0
0 500 1000 1500 2000 2500
Size in feet2
Supervised Learning Regression: Predict
“right answers” given continuous valued
output (price)
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.
Unsupervised Learning
Organize computing clusters Social network analysis
Image credit: NASA/JPL-Caltech/E. Churchwell (Univ. of Wisconsin, Madison)
Market segmentation Astronomical data analysis
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.
(c) Alexander Ihler