Lecture-1

Introduction to Computational
Intelligence
Dr. Abdul Majid, DCIS

1
 Intelligence

• A very general mental capability that, among

other things, involves the ability to reason,
plan, solve problems, think abstractly,
comprehend complex ideas, learn quickly and
learn from experience. [1]
• Measurement of Intelligence: IQ

2
 How do we think? What makes us intelligent?

• Cognition (The psychological result

of perception and learning and
reasoning) and Cognitive Modeling
– How do we think or how are we
intelligent?
• The Blue Brain Project [1]
– A biologically accurate,
functional model of the brain
using IBM's Blue Gene
supercomputer a biologically
accurate, functional model of
the brain using IBM's Blue Gene
supercomputer
[1] http://bluebrain.epfl.ch/

3
 Artificial and Computational
Intelligence
• Artificial and Computational Intelligence takes the problem of
understanding how we think a step further
– It attempts not just to understand it – but – also to build
intelligent entities
• A more proper definition of Artificial Intelligence
– The art of creating machines that perform functions that
require intelligence when performed by people [1]
• Measurement of Artificial Intelligence: Turing Test

[1] Rich E., and Knight K., (1991). Artificial Intelligence (2e), McGraw-Hill, NewYork.

4
 Artificial and Computational
Intelligence
• Application of the Turing Test
– CAPTCHA: Completely
Automated Public Turing test
to tell Computers and Humans
Apart [1]

[1] http://en.wikipedia.org/wiki/CAPTCHA
[2] http://en.wikipedia.org/wiki/List_of_Chatterbots
5
 Is ACI Intelligent?

• Planning
– Kasparov Vs. IBM Blue
– Time Table Schedulers

• Calculation
– Symbolic Integration in
Mathematica
– Theorem Provers

6
 Is ACI Intelligent?...

• Learning without a
Teacher
– ERDAS Imagine –
Classification of Land
Use
– Associative Memories
for data storage in
Databases

7
 Is ACI Intelligent?...

• Learning with a teacher

– No Hands Across America!
– Optical Character Recognition

Two researcher from CMU's Robotics Institute "drove" from

Pittsburgh, PA to San Diego using the RALPH computer
program. RALPH (Rapidly Adapting Lateral Position
Handler) uses video images to determine the location of the
road ahead and the appropriate steering direction to keep the
vehicle on the road
 Is ACI Intelligent?...
• Prediction
– Sunspot Number Prediction (Solar Cycle Prediction)

9
 Moving Towards Computational Intelligence…

• Classical AI (GOFAI: Good Old Fashioned AI)

– Concerned primarily with symbolic problems:
• Search Algorithms: e.g., A* Algorithm, Min-Max
• Knowledge Representation and Reasoning on
Knowledge base upon crisp logic
• Tarzan and Jane…
• Typical Example: SHRDLU

10
 An example of GOFAI in action:
Reasoning
• Reasoning based on Crisp Logic
• Definition of being tall in First Order Logic
– x Tall(x)  (Height(x) > 6.0)
• If Don has a height of 6.2
– Height(Don)=6.2
– The reasoning algorithm will fire the above defined rule
from its rule base to tell
• Tall(Don) = ??
• But what about
– Height(Sam)=5.11
• Tall(Sam) = ?
11
 Another Example of GOFAI: 8-
Queens
• Searching with backtracking
• Place N queens on a NxN chess board so
that no queen attacks any other
– Start with an empty board
– Place a queen in the first available
square
– Then place the next queen in the next
column so that the two queens do
not attack
– Keep on doing until either the
solution is reached or a queen starts
to attack another in which case back
track and choose another placement
option

12
 Moving Towards Computational Intelligence…

• Computational intelligence research aims to

use learning, adaptive, or evolutionary
algorithms to create programs that are, in
some sense, intelligent.

13
 Motivation for CI
• Fuzzy, Imprecise or Imperfect data
– Example: Imagine Designing a Car
Driving System and its rules being
taken from an illiterate driver!!!
– Expert opinions are often fuzzy

• No available mathematical relation

– Example: Finding the maxima of a
discontinuous unknown function!!
– Search spaces in practical can often be
discontinuous

14
 Motivation for CI
• Only Domain Experts are
Available
– Example: Medical Diagnosis
- No precise rules available
– Trigger for data mining
• Robustness and Adaptability
– Required in practical
scenarios
– Example: Pose invariant and
Lighting Invariant face
detection

15
 Motivation for CI
• Large Problems
– If the scale of a problem is large then GOFAI
algorithms have a difficult time dealing with them
• Solving N-Queens problem on a 256x256 chess board
– Here CI techniques offer better alternatives

16
 Inspirations for CI
• The previous desired characteristics are
exhibited and applied by intelligent living
beings
• So a good algorithm can be achieved by
considering how the minds of intelligent
beings evolve and work

17
 Biological Inspirations
• Artificial Neural Networks
– Mimic the learning process of the biological neuron

• Applications
– Disease Classification
– EEG Based Classification
– Load Forecasting
– Face Recognition
– No Hands Across America!
18
 Psychological Inspirations

• Our daily life descriptions of things is fuzzy

– Taller, smaller, heavier etc.
– So researchers have developed fuzzy systems
that provide effective solutions to complex
problems by mimicking this capability
• Applications
– EEG Based Classification
– Disease Diagnosis
– Control Applications
19
 Biological Inspirations
• Theory of Evolution
– Has Resulted in the Development of Genetic
Algorithms and Genetic Programming
– Multi-objective Genetic Algorithms
• Applications:
– Robotic Path Planning
– Automatic Program Building

20
 Behavioral Inspirations
• Swarm Intelligence
– Ant Colony Optimization
– Particle Swarm Optimization

21
 Course Objectives
• Focus Upon
– Artificial Neural Networks
– Genetic Algorithms,
– Genetic Programming
– Differential Evolution &
– Fuzzy Logic
– Ensemble Systems
• Evaluation
– Assignments
– Quizzes
– Tests
22