Professional Documents
Culture Documents
Artificial Intelligence: Artificial Intelligence - Introduction: AI Course Lecture 1-6, Notes, Slides
Artificial Intelligence: Artificial Intelligence - Introduction: AI Course Lecture 1-6, Notes, Slides
.in
rs
de
ea
yr
Artificial Intelligence - Introduction : AI Course Lecture 1- 6, notes, slides
.m
w
w
www.myreaders.info/ , RC Chakraborty, e-mail rcchak@gmail.com , June 01, 2010
,w
ty
www.myreaders.info/html/artificial_intelligence.html
or
ab
www.myreaders.info
kr
ha
C
C
R
Introduction
Artificial Intelligence
Introduction to Artificial Intelligence, topics : Definitions, goals,
approaches, techniques, and branches; Intelligent behavior,
understanding AI, hard or strong AI, soft or weak AI, cognitive
science. General, engineering and science based AI Goals. AI
approaches - cognitive science, laws of thought, turing test, rational
agent. AI Techniques that make system to behave as intelligent -
describe and match, goal reduction, constraint satisfaction, tree
searching, generate and test, rule based systems. Biology-inspired AI
techniques - neural networks, genetic algorithms, reinforcement
learning. Branches of AI - logical AI, search in AI, pattern recognition,
knowledge representation, inferencing, common sense knowledge and
reasoning, learning, planning, epistemology, ontology, heuristics,
genetic programming. Applications of AI - game playing, speech
recognition, understanding natural language, computer vision, expert
systems.
fo
.in
rs
de
ea
Introduction
yr
.m
w
w
,w
Artificial Intelligence
ty
or
ab
kr
ha
Topics
C
C
R
1. Definitions 03-10
2. Goals of AI 11-12
3. AI Approaches 13-16
4. AI Techniques 17-32
5. Branches of AI 33-45
6. Applications of AI 46-50
7. References 51
02
fo
.in
rs
de
ea
yr
Introduction
.m
w
w
,w
ty
ea
1. Definitions
yr
.m
w
w
1.1 Artificial Intelligence (AI)
,w
ty
or
ab
(a) 'The exciting new effort to make (b) 'The study of mental faculties through
R
(C) 'The art of creating machines that (d) 'A field of study that seeks to explain
perform functions that require and emulate intelligent behavior in
intelligence when performed by terms of computational processes'
people' (Kurzweil, 1990) (Schalkoff, 1990)
'The study of how to make computers 'The branch of computer science that
do things at which, at the moment, is concerned with the automation of
people are better' (Rich and Knight, intelligent behavior' (Luger and
1991) Stubblefield, 1993)
ea
1.2 Intelligence
yr
.m
w
w
,w
Examples:
ab
kr
ha
05
fo
.in
rs
de
AI - Definitions
ea
1.3 Intelligent Behavior
yr
.m
w
w
,w
06
fo
.in
rs
de
AI - Definitions
ea
1.4 Understanding AI
yr
.m
w
w
,w
07
fo
.in
rs
de
AI - Definitions
ea
1.5 Hard or Strong AI
yr
.m
w
w
,w
ea
1.6 Soft or Weak AI
yr
.m
w
w
,w
ea
1.7 Cognitive Science
yr
.m
w
w
,w
▪ Example : The Chess programs are successful, but say little about
the ways humans play chess.
10
fo
.in
rs
de
Goals of AI
ea
2. Goals of AI
yr
.m
w
w
,w
• Traditionally, all four goals have been followed and the approaches were:
Human-like Rationally
ea
• General AI Goal
yr
.m
w
w
Replicate human intelligence : still a distant goal.
,w
ty
or
ea
3. AI Approaches
yr
.m
w
w
The approaches followed are defined by choosing goals of the computational
,w
ty
or
ea
3.2 Laws of Thought : Think Rationally
yr
.m
w
w
▪ The study of mental faculties through the use of computational models;
,w
ty
or
▪ The issue is, not all problems can be solved just by reasoning
and inferences.
14
fo
.in
rs
de
AI Approaches
ea
3.3 Turing Test : Act Human-like
yr
.m
w
w
▪ The art of creating machines that perform functions requiring
,w
ty
or
ea
3.4 Rational Agent : Act Rationally
yr
.m
w
w
▪ Tries to explain and emulate intelligent behavior in terms of
,w
ty
or
intelligence.
ha
C
C
R
16
fo
.in
rs
de
AI Techniques
ea
4. AI Techniques
yr
.m
w
w
Various techniques that have evolved, can be applied to a variety of AI tasks.
,w
ty
or
The techniques are concerned with how we represent, manipulate and reason
ab
kr
Example
17
fo
.in
rs
de
AI Techniques
ea
4.1 Techniques that make system to behave as "Intelligent"
yr
.m
w
w
,w
■ Finite state model consists of a set of states, a set of input events and
R
the relations between them. Given a current state and an input event
you can determine the next current state of the model.
■ Transition relation: If a pair of states (S, S') is such that one move
takes the system from S to S', then the transition relation is represented
by S => S’
■ Examples of some possible transitions between states are shown for the
Towers of Hanoi puzzle.
18
fo
.in
rs
de
AI Techniques
ea
■ Puzzle : Towers of Hanoi with only 2 disks
yr
.m
w
w
,w
Move the disks from the leftmost post to the rightmost post while
never putting a larger disk on top of a smaller one;
move one disk at a time, from one peg to another;
middle post can be used for intermediate storage.
Play the game in the smallest number of moves possible .
[1, 2] [ ] []
[ 2] [1 ] [] [2] [] [1]
19
fo
.in
rs
de
AI Techniques
ea
• Goal Reduction
yr
.m
w
w
,w
based representations.
C
C
R
20
fo
.in
rs
de
AI Techniques
ea
■ Example Goal Reduction
yr
.m
w
w
AND-OR tree/graph structure to represent facts such as “enjoyment”,
,w
ty
or
OR
OR
Improve productivity
Go on strike
◊ Conjoint sub-goals
To “provide for old age”, not only need to “earn more money”, but
as well need to “save money”.
21
fo
.in
rs
de
AI Techniques
ea
• Constraint Satisfaction Techniques
yr
.m
w
w
,w
ea
[continued from Constraint Satisfaction Techniques]
yr
.m
w
w
■ Constraint satisfaction has application in Artificial Intelligence,
,w
a b c d e f g h a b c d e f g h
8 8 8 8
7 7 7 7
6 6 6 6
5 5 5 5
4 4 4 4
3 3 3 3
2 2 2 2
1 1 1 1
a b c d e f g h a b c d e f g h
Unique solution 1 Unique solution 2
ea
■ Example 2 : Map Coloring
yr
.m
w
w
Problem : Given a map (graph) and a number of colors, the problem is
,w
ty
or
constraint that no adjacent nodes (areas) have the same color assigned
ha
C
C
to them.
R
ea
• Tree Searching
yr
.m
w
w
■ Many problems (e.g. goal reduction, constraint networks) can be
,w
ty
or
◊ Depth-first search
* In the case of a dead end - backtrack to the last node in the tree
where a previously unexplored path branches of, and test this path.
* Backtracking can be of two types :
– Chronological backtracking : undo everything as we move back
"up" the tree to a suitable node.
– Dependency directed backtracking : only withdraw choices that
"matter" (ie those on which dead end depends).
25
fo
.in
rs
de
AI Techniques
ea
• Generate and Test (GT)
yr
.m
w
w
■ Most algorithms for solving Constrain Satisfaction Problems (CSPs)
,w
ty
or
■ Generate-and-test method
The method first guesses the solution and then tests whether this
solution is correct, means solution satisfies the constraints.
◊ This paradigm involves two processes:
* Generator to enumerate possible solutions (hypotheses).
* Test to evaluate each proposed solution
◊ The algorithm is
Generate labels
Test satisfaction
Disadvantages
ea
• Rule-Based Systems (RBSs)
yr
.m
w
w
■ Rule-based systems are simple and successful AI technique.
,w
ty
or
ab
Change Interpreter
Conditions
Observed Rules
Working
Data Memory
Rule
Base
Rule-Based System
ea
■ RBS components - Description
yr
.m
w
w
◊ Working Memory (WM)
,w
ty
or
e.g. < car, color, red > : “The color of my car is red”.
R
◊ Interpreter
‡ It is the domain independent reasoning mechanism for RBS.
‡ It selects rule from Rule Base and applies by performing action.
‡ It operates on a cycle:
ea
4.2 Biology-Inspired AI Techniques
yr
.m
w
w
• Neural Networks (NN)
,w
ty
or
ab
ea
• Genetic Algorithms (GA)
yr
.m
w
w
■ GAs are part of evolutionary computing, a rapidly growing area of AI.
,w
ty
or
ab
◊ When two organisms mate they share their genes. The resultant
offspring may end up having half the genes from one parent and
half from the other. This process is called cross over.
◊ A gene may be mutated and expressed in the organism as a
completely new trait.
ea
■ Genetic Algorithm Steps
yr
.m
w
w
(1) [Start] Generate random population of n chromosomes (Encode
,w
ty
or
population.
R
(4) [Replace] Use new generated population for a further run of the
algorithm.
(5) [Test] If the end condition is satisfied, stop, and return the best
ea
• Reinforcement Learning (RL)
yr
.m
w
w
■ RL is learning from interaction with an environment; from the
,w
ty
or
processes (MDPs).
R
maximize the cumulative reward it receives over time. The steps are:
◊ At each time t, the agent perceives its environment state st and
the set of possible actions A(st) .
◊ It chooses an action a ∈ A(st) and receives from the environment
the new state st+1 and a reward rt+1.
◊ Based on these interactions, the agent develops a policy π : S A
ea
5. Branches of AI
yr
.m
w
w
• Logical AI
,w
ty
or
ab
■ Types of logic
◊ Propositional logic - logic of sentences
◊ predicate logic - logic of objects
◊ logic involving uncertainties
◊ Fuzzy logic - dealing with fuzziness
◊ Temporal logic, etc
■ Propositional logic and Predicate logic are fundamental to all logic
◊ Propositional logic
‡ Propositions are “Sentences”; either true or false but not both.
‡ A sentence is smallest unit in propositional logic
‡ If proposition is true, then truth value is "true"; else “false”
‡ Example : Sentence "Grass is green"; Truth value “ true”;
Proposition is “yes”
◊ Predicate logic
‡ Predicate is a function may be true or false for arguments
‡ Predicate logic are rules that govern quantifiers
‡ Predicate logic is propositional logic added with quantifiers
‡ Examples:
“The car Tom is driving is blue",
"The sky is blue",
"The cover of this book is blue"
Predicate is blue, give a name B ;
Sentence represented as B(x); read B(x) as "x is blue" ;
Object represented as x .
33
fo
.in
rs
de
Branches of AI
ea
• Search in AI
yr
.m
w
w
■ Search is a problem-solving technique that systematically consider
,w
ty
or
all possible action to find a path from initial state to target state.
ab
kr
ha
C
■ Search components
◊ Initial state - First location
◊ Available actions - Successor function : reachable states
◊ Goal test - Conditions for goal satisfaction
◊ Path cost - Cost of sequence from initial state to reachable state
■ Search objective
◊ Transform initial state into goal state - find a sequence of actions.
■ Search solution
◊ Path from initial state to goal - optimal if lowest cost.
34
fo
.in
rs
de
Branches of AI
ea
• Pattern Recognition (PR)
yr
.m
w
w
■ Definitions : from the literature
,w
ty
or
ab
◊ Template Matching
◊ Statistical classification
ea
[continued from previous slide Approaches for Pattern Recognition ]
yr
.m
w Template Matching
w
,w
ty
Statistical classification
Each pattern is represented in terms of d features (measurements) and
viewed as a point in a d-dimensional space. Using training sets
establish decision boundaries in the feature space - following decision
theoretic or discriminant analysis approaches.
Neural networks
Neural networks are viewed as weighted directed graphs in which the
nodes are artificial neurons and directed edges (with weights) are
connections between neurons input-output. Neural networks have the
ability to learn complex nonlinear input-output relationships from the
sequential training procedures, and adapt themselves to input data.
ea
• Knowledge Representation
yr
.m
w
w
■ How do we represent what we know?
,w
ty
or
ea
• Inference
yr
.m
w
w
Inference is the act or process of deriving a conclusion based solely on what
,w
ty
or
one already knows; it is deduction of new facts from old ones; Logic
ab
kr
captures inference.
ha
C
C
R
■ Deductive Inference
■ Inductive Inference
ea
• Common Sense Knowledge and Reasoning
yr
.m
w
w
Common sense is the mental skills that most people have.
,w
ty
or
ab
ea
• Learning
yr
.m
w
w
Programs learn from what the facts or the behaviors can represent.
,w
ty
or
ab
■ Definitions
kr
ha
that are adaptive in the sense that they enable the system to do the
same task or tasks more efficiently and more effectively the next
time.”
◊ Marvin Minsky 1986 – “Learning is making useful changes in the
working of our mind.”
◊ Ryszard Michalski 1986 – "Learning is constructing or modifying
representations of what is being experienced."
◊ Mitchell 1997 – “A computer program is said to learn from
experience E with respect to some class of tasks T and performance
measure P, if its performance at tasks in T, as measured by P,
improves with experience E.”
ea
• Planning
yr
.m
w
w
A plan is a representation of a course of action.
,w
ty
or
■ Planning programs
Start with facts about the world, particularly
◊ facts about the effects of actions,
◊ facts about the particular situation, and
◊ statement of a goal.
■ Benefits of planning
◊ reducing search,
◊ resolving goal conflicts, and
◊ providing a basis for error recovery.
ea
• Epistemology
yr
.m
w
w
Epistemology is the theory of knowledge.
,w
ty
or
ab
Propositions
Know-
Truths ledge Beliefs
ea
• Ontology
yr
.m
w
w
Ontology is concerned with existence; a study of the categories of things
,w
ty
or
about the objects in that domain and the relations between them.
ea
• Heuristics
yr
.m
w
w
Heuristics are simple, efficient rules;
,w
ty
or
ab
◊ used to compare two nodes, find if one is better than the other.
44
fo
.in
rs
de
Branches of AI
ea
• Genetic programming (GP)
yr
.m
w
w
Genetic programming is an automated method for creating program
,w
ty
or
■ The major five preparatory steps, the human user require to specify
ea
6. Applications of AI
yr
.m
w
w
• Game playing
,w
ty
or
ab
e.g. enemies, partners, and support characters that act just like
humans.
◊ "Unpredictable" opponent
◊ Need to specify move for every possible opponent reply.
◊ Time limits - games become boring if there is no action for too long
a time; opponents are unlikely to find goal, must approximate.
■ Computer Games
■ The Deep Blue Chess program won over world champion Gary
Kasparov.
46
fo
.in
rs
de
Applications of AI
ea
• Speech Recognition
yr
.m
w
w
■ A process of converting a speech signal to a sequence of words;
,w
ty
or
ab
limited purposes.
C
C
R
ea
• Understanding Natural Language
yr
.m
w
w
Natural language processing (NLP) does automated generation and
,w
ty
or
ea
• Computer Vision
yr
.m
w
w
■ It is a combination of concepts, techniques and ideas from : Digital
,w
ty
or
Computer Graphics.
ha
C
C
R
■ The world is composed of 3-D objects, but the inputs to the human eye
■ Some useful programs can work solely in 2-D, but full computer vision
requires partial 3-D information that is not just a set of 2-D views.
directly, and they are not as good as what humans evidently use.
■ Examples
◊ Face recognition :
the programs in use by banks
◊ Autonomous driving :
The ALVINN system, autonomously drove a van from Washington,
D.C. to San Diego, averaging 63 mph day and night, and in all
weather conditions.
◊ Other usages
Handwriting recognition, Baggage inspection, Manufacturing
inspection, Photo interpretation, etc .
49
fo
.in
rs
de
Applications of AI
ea
• Expert Systems
yr
.m
w
w
Systems in which human expertise is held in the form of rules
,w
ty
or
ab
being present.
C
R
■ Knowledge base
A knowledge engineer interviews experts in a certain domain and tries
to embody their knowledge in a computer program for carrying out
some task.
■ One of the first expert systems was MYCIN in 1974, which diagnosed
ea
7. References : Textbooks
yr
.m
w
w
1. "Artificial Intelligence", by Elaine Rich and Kevin Knight, (2006), McGraw Hill
,w
ty
5. "AI: A New Synthesis", by Nils J. Nilsson, (1998), Morgan Kaufmann Inc., Chapter
1-25, Page 1-493.
51