UNIT-I

Syllabus
Introduction: What is AI, Foundations of AI, History of AI, The State of Art.

Intelligent Agents: Agents and Environments, Good Behavior: The Concept of Rationality,
The Nature of Environments, The Structure of Agents.

2
 CHAPTER-1 INTRODUCTION
The field of artificial intelligence, or AI, goes further still: it attempts not just to understand but also
to build intelligent entities.

AI is one of the newest sciences. Work started in earnest soon after World War II, and the name
itself was coined in 1956.

Along with molecular biology, AI is regularly cited as the "field I would most like to be in" by
scientists in other disciplines.

A student in physics might reasonably feel that all the good ideas have already been taken by Galileo,
Newton, Einstein, and the rest. AI, on the other hand, still has openings for several full-time
Einsteins.

AI systematizes and automates intellectual tasks and is therefore potentially relevant to any sphere of
human intellectual activity. In this sense, it is truly a universal field.

1.1.1. What is AI?

Views of AI fall into four categories:

Note: A system is rational if it does the “right

A. Acting Humanly : The Turing Test Approach

Alan Turing (1950) - “Computing machinery and intelligence":

Can machines think? Can machines behave intelligently?
Operational test for intelligent behavior: the Imitation Game.

3
 Predicted that by 2000, a machine might have a 30% chance of fooling a lay person for 5 minutes.
Anticipated all major arguments against AI in following 50 years.
Suggested major components of AI: knowledge, reasoning, language understanding, learning, computer
vision & Robotics.
 Problem: Turing test is not reproducible, constructive, or amenable to mathematical analysis.

B. Thinking humanly: The Cognitive modeling approach

1960s - cognitive revolution": information processing psychology replaced prevailing orthodoxy of

behaviorism”.
Requires scientific theories of internal activities of the brain.
What level of abstraction? Knowledge" or circuits"?
 What is cognitive science??
The study of thought, learning, and mental organization, which draws on aspects of psychology,
linguistics, philosophy, and computer modeling.
 How to validate? Requires
1) Predicting and testing behavior of human subjects (top-down) or
2) Direct identification from neurological data (bottom-up).
Both approaches (roughly, Cognitive Science and Cognitive Neuroscience) are now distinct from AI

C. Thinking rationally : The “laws of thought” approach

Normative (or prescriptive) rather than descriptive

Aristotle: what are correct arguments/thought processes?
Several Greek schools developed various forms of logic: notation and rules of derivation for thoughts;
May or may not have proceeded to the idea of mechanization.
Direct line through mathematics and philosophy to modern AI
 Problems:
1) Not all intelligent behavior is mediated by logical deliberation
2) What is the purpose of thinking? What thoughts should I have?

D. Acting rationally : The rational agent approach

Rational behavior: doing the right thing
The right thing: that which is expected to maximize goal achievement, given the available information.
Doesn't necessarily involve thinking e.g., blinking reflex, but thinking should be in the service of
4
 Aristotle (Nicomachean Ethics): Every art and every inquiry, and similarly every action and pursuit, is
thought to aim at some good.

1.1.2. Foundations of AI…

The different foundations which are contributors for AI are:
i. Philosophy
ii. Mathematics
iii. Economics
iv. Neuroscience
v. Psychology
vi. Computer engineering
vii.Control theory and cybernetics
viii.Linguistics

i. Philosophy(428 B .C .-present):
Can formal rules be used to draw valid conclusions?
How does the mental mind arise from a physical brain?
Where does knowledge come from?
How does knowledge lead to action?
Aristotle (384-322 B.C.) was the first to formulate a precise set of laws governing the rational part of the
mind.

ii. Mathematics(B.C 800-present):

What are the formal rules to draw valid conclusions?
What can be computed?
How do we reason with uncertain information?
Besides logic and computation, the third great contribution of mathematics to A1 is PROBABILITY - the
theory of probability.
The Italian Gerolamo Cardano (1501-1576) first framed the idea of probability, describing it in terms of the
possible outcomes of gambling events.
Thomas Bayes (1702-1 761) proposed a rule for updating probabilities in the light of new evidence.
Bayes' rule and the resulting field called Bayesian analysis form the basis of most modern approaches to
uncertain reasoning in A1 systems.

iii. Economics (1776-present):

How should we make decisions so as to maximize payoff?
How should we do this when others may not go along?
How should we do this when the payoff may be fix in the future?
Work in economics and operations research has contributed much to our notion of rational Agents.
Herbert Simon (191 6- 2001), the pioneering AI researcher, won the Nobel prize in economics in 1978 for his
early SATISFICING work showing that models based on satisficing-making decisions that are "good
5
 enough“.

iv Neuroscience (1861-present):

How do brains process information?

Neuroscience is the study of the nervous system, particularly the brain.
Paul Broca's (1824-1880) study of aphasia (speech deficit) in brain-damaged patients.
He proved speech production was localized to a portion of the left hemisphere now called NEURONS Broca's
area.
Carnillo Golgi (1843-1926) developed a staining technique allowing the observation of individual neurons in
the brain.
The recent development of functional magnetic resonance imaging (MRI) (Ogawa et al., 1990) is giving
neuroscientists unprecedentedly detailed images of brain activity.

 NOTE :Moore’s Law predicts that the CPU’s gate count will equal the brain’s neuron count around
2020.

6
v Psychology (1879-present):

How do humans and animals think and act?

The origin of scientific psychology are traced back to the wok if German physiologist Hermann von
Helmholtz(1821-1894) and his student Wilhelm Wundt(1832 – 1920).
In 1879,Wundt opened the first laboratory of experimental psychology at the university of Leipzig.
In US, the development of computer modeling led to the creation of the field of cognitive science.
The field can be said to have started at the workshop in September 1956 at MIT.
vi Computer Engineering (1940-present):

How can we build an efficient computer?

For artificial intelligence to succeed, we need two things: intelligence and an artifact.
The computer has been the artifact of choice.
AI also owes a debt to the software side of computer science, which has supplied the operating systems,
programming languages, and tools needed to write modern programs.

vii Control theory and Cybernetics (1948-present):

How can artifacts operate under their own control?

Ktesibios of Alexandria (c. 250 B.C.) built the first self-controlling machine: a water clock.
It contains regulator that kept the flow of water running through it at a constant, predictable pace.
Modern control theory, especially the branch known as stochastic optimal control, has as its goal the design of
systems that maximize an objective function over time.

Viii Linguistics:

How does language relate to thought?

Modem linguistics and AI, then, were "born" at about the same time, and grew up together, intersecting in a
hybrid field called computational linguistics or natural language processing.

1.1.3. History of AI...

A. Gestation, 1943–56:

McCulloch & Pitts (1943)

Artificial neural net—proved equivalent to Turing machine;
Shannon, Turing (1950)
Chess playing programs
Marvin Minsky (1951)
First neural net computer—SNARC
Dartmouth College (1956)
Term “AI” coined by John McCarthy
Newell & Simon presented LOGIC THEORIST program

7
B. Early enthusiasm, 1952–69:
Lots of progress.
Programs written that could:
– plan, learn
– play games,
– prove theorems, in general, solve problems.
Major feature of the period were microworlds—toy problem domains.

Example: blocks world.

Planners simulated manipulating simple shapes like this:

Even some simple natural language handling:

8
 C. A dose of reality, 1966–1973:

Techniques developed on microworlds would not scale.

Implications of complexity theory—developed in late 1960s, early 1970s—began to be appreciated:
o brute force techniques will not work.
o works in principle does not mean works in practice.
Lots of early programs did symbol manipulation without any real understanding of the domain they were in.

D. Knowledge based systems, 1969–1979:

General purpose, brute force techniques don’t work, so use knowledge rich solutions.
Early 1970s saw emergence of expert systems as systems capable of exploiting knowledge about tightly
focused domains to solve problems normally considered the domain of experts.
Ed Feigenbaum’s knowledge principle: [Knowledge] is power, and computers that amplify that knowledge
will amplify every dimension of power.
Expert systems success stories:
o MYCIN — blood diseases in humans;
o DENDRAL — interpreting mass spectrometers;
o R1/XCON — configuring DEC VAX hardware;
o PROSPECTOR —finding promising sites for mineral deposits;
Expert systems emphasised knowledge representation: rules, frames, semantic nets.
 Problems:
– the knowledge elicitation bottleneck;
– marrying expert system & traditional software;
– breaking into the mainstream.

E. AI makes money, 1980–present:

R1 was the first commercial expert system.
Led to a boom in expert systems companies.
9
 o Like an earlier dot com boom
Most of those companies failed to deliver increased productivity for their customers.
They went bust.
AI suffered from all the broken promises, but didn’t go away.

F. AI and the scientific method, 1987– present:

Connection back to other fields.

o Probability/statistics
o Control theory
o Economics
o Operations research.
New industries spawned.
o Data mining is machine learning.
Low-key successes.
o Expert systems embedded in Windows.
Increased emphasis on shared datasets, common problems, competitions.

G. Intelligent agents, 1993–present:

Emphasis on understanding the interaction between agents and environments.

AI as component, rather than as end in itself.
o “Useful first” paradigm — Etzioni (NETBOT, US$35m)
o “Raisin bread” model — Winston.
More about interaction between components, emergent intelligence, and doing well enough to be useful.

10
H. Large datasets, 2001–present:

Change in emphasis from algorithm to data.

o Perhaps with enough data, we don’t have to be smart to be good.
For example, how to fill in a gap in a picture.
o Perhaps you have a picture of your ex in front of a great landscape.
Poll lots of similar pictures for a matching piece.
o Moving from 10,000 images to 2,000,000 led to a big jump in performance.
Wisdom of the crowd.
The growth of the internet makes this much easier in 2010 than it was in 1990.

Currently a lot of interest in mining data from social networks.

If people are connected, they are similar in some sense.

1.1.4. THE STATE OF THE ART:

Here we sample a few applications;
1. Robotic vehicles: A driverless robotic car named STANLEY sped through the rough terrain of the
Mojave dessert at 22 mph, finishing the 132-mile course first to win the 2005 DARPA Grand
Challenge.
2. Speech recognition: A traveler calling United Airlines to book a flight can have the entire
conversation guided by an automated speech recognition and dialog management system.
3. Autonomous planning and scheduling: A hundred million miles from Earth, NASA’s Remote
Agent program became the first on-board autonomous planning program to control the scheduling of
operations for a spacecraft.
4. Game playing: IBM’s DEEP BLUE became the first computer program to defeat the world champion in a
chess match when it bested Garry Kasparov by a score of 3.5 to 2.5 in an exhibition match.
5. Spam fighting: Each day, learning algorithms classify over a billion messages as spam, saving the recipient
from having to waste time deleting what, for many users, could comprise 80% or 90% of all messages, if not
classified away by algorithms.
6. Logistics planning: During the Persian Gulf crisis of 1991, U.S. forces deployed a Dynamic Analysis and
Replanning Tool, DART (Cross and Walker, 1994), to do automated logistics planning and scheduling for
transportation.
7. Robotics: The iRobot Corporation has sold over two million Roomba robotic vacuum cleaners for home use.
8. Machine Translation: A computer program automatically translates from Arabic to English .

11
 UNIT-I CHAPTER -2 INTELLIGENT AGENTS

1.2.1. AGENTS AND ENVIRONMENT :

 Agent Definition…

An agent is a system that:

– is situated in an environment,
– is capable of perceiving its environment, and
– is capable of acting in its environment with the goal of satisfying its design objectives.
(or)

Anything that can be viewed as perceiving its environment through sensors and SENSOR acting upon that
environment through actuators.

(OR)

Figure : An agent example representation

12
  Examples for different agents:

i. Human “agent”:
– environment: physical world;
– sensors: eyes, ears, . . .
– effectors: hands, legs, . . .
ii. Software agent:
– environment: (e.g.) UNIX operating system;
– sensors: keystrokes, file contents, n/w packets . . .
– effectors: rm, chmod, . . .
iii. Robot:
– environment: physical world;
– sensors: sonar, camera;
– effectors: wheels.

 Important terms…
Percept:
We use the term percept to refer to the agent's perceptual inputs at any given instant.
Percept Sequence:
An agent's percept sequence is the complete history of everything the agent has ever perceived.
Agent function: f:P* →A
Mathematically speaking, we say that an agent's behavior is described by the agent function that maps any given
percept sequence to an action.

 Agent function VS Agent program:

The agent function is an abstract mathematical description; the agent program is a concrete implementation, running
on the agent architecture.
Example: The vacuum-cleaner world…

13
 1.2.2 Good behavior – Concept of Rationality…
 Rational Agent…

A rational agent is one that does the right thing-conceptually speaking, every entry in the table for the agent function
is filled out correctly.

Obviously, doing the right thing is better than doing the wrong thing. The right action is the one that will cause the
agent to be most successful.

 Performance measures:

A performance measure embodies the criterion for success of an agent's behavior.

When an agent is plunked down in an environment, it generates a sequence of actions according to the percepts it
receives.
This sequence of actions causes the environment to go through a sequence of states. If the sequence is
desirable, then the agent has performed well.

Rationality :

What is rational at any given time depends on four things:

The performance measure that defines the criterion of success.
The agent's prior knowledge of the environment.
The actions that the agent can perform.
The agent's percept sequence to date.

Omniscience, learning, and autonomy…

An omniscient agent knows the actual outcome of its actions and can act accordingly; but omniscience is
impossible in reality.
Doing actions in order to modify future percepts-sometimes called information gathering-is an important part
of rationality.
Our definition requires a rational agent not only to gather information, but also to learn as much as possible
from what it perceives.
To the extent that an agent relies on the prior knowledge of its designer rather than on its own percepts, we
say that the agent lacks autonomy. A rational agent should be autonomous-it should learn what it can to
compensate for partial or incorrect prior knowledge.

1.2.3 The Nature of Environments :

Task environments :
These are essentially the "problems" to which rational agents are the "solutions.“
Specifying the task environment : PEAS
The rationality of the simple vacuum-cleaner agent, needs specification of …
– the performance measure - P
– the environment - E
– the agent's actuators and sensors - A & S
14
PEAS Example…

Properties of task environments…

I. Fully observable vs. partially observable
II. Deterministic vs. stochastic
III. Episodic vs. sequential
IV. Static vs. dynamic
V. Discrete vs. continuous
VI. Single agent vs. Multiagent

15
I. Fully observable vs. partially observable…
If an agent's sensors give it access to the complete state of the environment at each point in time, then we say
that the task environment is fully observable.
A task environment is effectively fully observable if the sensors detect all aspects that are relevant to the
choice of action;
An environment might be partially observable because of noisy and inaccurate sensors or because parts of the
state are simply missing from the sensor.

II. Deterministic vs. stochastic…

If the next state of the environment is completely determined by the current state and the action executed by the
agent, then we say the environment is deterministic; other-wise, it is stochastic.

III. Episodic vs. sequential…

In an episodic task environment, the agent's experience is divided into atomic episodes.
Each episode consists of the agent perceiving and then performing a single action.
Crucially, the next episode does not depend on the actions taken in previous episodes.
In sequential environments, on the other hand, the current decision could affect all future decisions. Chess and
taxi driving are sequential.

IV. Static vs. Dynamic...

If the environment can change while an agent is deliberating, then we say the environment is dynamic for that
agent; otherwise, it is static.
Static environments are easy to deal with because the agent need not keep looking at the world while it is
deciding on an action, nor need it worry about the passage of time.
Dynamic environments, on the other hand, are continuously asking the agent what it wants to do; if it hasn’t
decided yet, that counts as deciding to do nothing.
If the environment itself does not change with the passage of time but the agent’s performance score does,
then we say the environment is semidynamic. Taxi driving is clearly dynamic: the other cars and the taxi
itself keep moving while the driving algorithm dithers about what to do next. Chess, when played with a
clock, is semidynamic. Crossword puzzles are static.

V. Discrete vs. continuous…

A discrete-state environment such as a chess game has a finite number of distinct states.
Chess also has a discrete set of percepts and actions.
Taxi driving is a continuous state and continuous-time problem: the speed and location of the taxi and of the
other
vehicles sweep through a range of continuous values and do so smoothly over time.
Taxi-driving actions are also continuous (steering angles,.)

VI. Single agent vs. multiagent…

An agent solving a crossword puzzle by itself is clearly in a single-agent environment, whereas an agent

16
 playing chess is in a two-agent environment.

As one might expect, the hardest case is partially observable, stochastic, sequential, dynamic, continuous, and
Multiagent.

1.2.4 The Structure of Agents…

 Agent programs:

The job of AI is to design the agent program that implements the agent function mapping percepts to actions.
Agent = architecture +program
The agent programs all have the same skeleton: they take the current percept as input from the sensors and
return an action to the actuators.

Notice the difference between the agent program, which takes the current percept as input, and the agent
function, which takes the entire percept history.

 Agent Types/Categories…
i. Table-driven agents:
-- use a percept sequence/action table in memory to find the next action. They are implemented by a (large) lookup
table.
ii. Simple reflex agents:
-- are based on condition-action rules, implemented with an appropriate production system. They are stateless devices
which do not have memory of past world states.
iii. Agents with memory (Model):
-- have internal state, which is used to keep track of past states of the world.
iv. Agents with goals:
-- are agents that, in addition to state information, have goal information that describes desirable situations. Agents of
this kind take future events into consideration.
v. Utility-based agents:
--base their decisions on classic axiomatic utility theory in order to act rationally.
vi. Learning Agents:

17
 i. Table Driven Agent:

Drawbacks:
Table lookup of percept-action pairs defining all possible condition-action rules necessary to interact in an
environment.
 Problems :
– Too big to generate and to store (Chess has about 10^120 states, for example)
– No knowledge of non-perceptual parts of the current state
– Not adaptive to changes in the environment; requires entire table to be updated if changes occur
– Looping: Can't make actions conditional
– Take a long time to build the table
– No autonomy
– Even with learning, need a long time to learn the table entries

ii. Simple Reflex Agent:

The simplest kind of agent is the simple reflex agent.

These agents select actions on the basis of the current percept, ignoring the rest of the percept history. E.g. the
vacuum-agent
Large reduction in possible percept/action situations.
Implemented through condition-action rules.
Simple reflex behaviors occur even in more complex environments. Imagine yourself as the driver of the
automated taxi. If the car in front brakes and its brake lights come on, then you should notice this and initiate
braking.
In other words, some processing is done on the visual input to establish the condition we call “The car in
front is braking.” Then, this triggers some established connection in the agent program to the action “initiate
braking.”
We call such a connection a condition–action rule,5 written as if car-in-front-is-braking then initiate-
braking.
Example : Vaccum Cleaner : If dirty then suck

18
 Characteristics:
Only works if the environment is fully observable.
Lacking history, easily get stuck in infinite loops.
One solution is to randomize actions.

iii. Model based reflex Agents:

The most effective way to handle partial observability is for the agent to keep track of the part of the world it
can't see now.

That is, the agent should maintain some sort of internal state that depends on the percept history and thereby
reflects at least some of the unobserved aspects of the current state.

Updating this internal state information as time goes by requires two kinds of knowledge to be encoded in the
19
 agent program.

First, we need some information about how the world evolves independently of the agent-for examp le,
that
an overtaking car generally will be closer behind than it was a moment ago.

Second, we need some information about how the agent's own actions affect the world-for example, that
when the agent turns the steering wheel clockwise, the car turns to the right or that after driving for five
minutes northbound on the freeway one is usually about five miles north of where one was five minutes
ago.

This knowledge about "how the world working -whether implemented in simple Boolean circuits or
in complete scientific theories-is called a model of the world. An agent that uses such a model is called a
model-based agent.

iv Goal Based Agents:

Knowing about the current state of the environment is not always enough to decide what to do. For example,
at a road junction, the taxi can turn left, turn right, or go straight on.

The correct decision depends on where the taxi is trying to get to.

In other words, as well as a current state description, the agent needs some sort of goal information that
describes situations that are desirable-for example, being at the passenger's destination.
20
 The agent program can combine this with information about the results of possible actions (the same
information as was used to update internal state in the reflex agent) in order to choose actions that achieve the
goal.

Sometimes goal-based action selection is straightforward—for example, when goal satisfaction results
immediately from a single action. Sometimes it will be more tricky—for example, when the agent has to
consider long sequences of twists and turns in order to find a way to achieve the goal.

Search and planning are the subfields of AI devoted to finding action sequences that achieve the agent’s
goals.

Although the goal-based agent appears less efficient, it is more flexible because the knowledge that supports
its decisions is represented explicitly and can be modified.

If it starts to rain, the agent can update its knowledge of how effectively its brakes will operate; this will
automatically cause all of the relevant behaviors to be altered to suit the new conditions.

V Utility based Agents:

Goals alone are not really enough to generate high-quality behavior in most environments. For example, there
are many action sequences that will get the taxi to its destination (thereby achieving the goal) but some are
quicker, safer, more reliable, or cheaper than others.
Goals just provide a crude binary distinction between "happy" and "unhappy" states, whereas a more general
performance measure should allow a comparison of different world states according to exactly how happy
they would make the agent if they could be achieved.

Because "happy" does not sound very scientific, the customary terminology is to say that if one world state is
preferred to another, then it has higher utility for the agent.

A utility function maps a state (or a sequence of states) onto a real number, which describes the associated
degree of happiness.
A complete specification of the utility function allows rational decisions in two kinds of cases where goals are
21
 inadequate.

First, when there are conflicting goals, only some of which can be achieved (for example, speed and safety), the
utility function specifies the appropriate tradeoff.
Second, when there are several goals that the agent can aim for, none of which can be achieved with certainty,
utility provides a way in which the likelihood of success can be weighed up against the importance of the
goals.
Partial observability and stochasticity are ubiquitous in the real world, and so, therefore, is decision making
under uncertainty.

Technically speaking, a rational utility-based agent chooses the action that maximizes the expected utility of
the action outcomes—that is, the utility the agent expects to derive, on average, given the probabilities and
utilities of each outcome.

Vi Learning Agents:

All agents can improve their performance through learning.

A learning agent can be divided into four conceptual components as shown in figure:
a. Learning element
b. Performance element
c. Critic
d. Problem generator

22
a. Learning element - is responsible for making improvements.
b. Performance element - is responsible for selecting external actions. The performance element is what we have
previously considered to be the entire agent: it takes in percepts and decides on actions.
c. Critic - The learning element uses feedback from the critic on how the agent is doing and determines how the
performance element should be modified to do better in the future.
d. Problem generator - It is responsible for suggesting actions that will lead to new and informative experiences.

*********************

23