Artificial Intelligence

It has seven chapters

Compiled by Yared Ab
 Chapter 1: Introduction to AI
 Chapter 2: Intelligent Agent
Artificial = man made  Chapter 3: Problem Solving
Intelligence = thinking by Searching
power  Chapter 4: Knowledge
Representation & Reasoning
 Chapter 5: Machine Learning
Basics
 Chapter 6: Natural
Language Processing
Basics
 Chapter 7: Robotic Sensing
and Manipulation
 Chapter 1

Introduction to Artificial
Intelligence
 Artificial intelligence
❖What is intelligence ?
 A general mental capability or property of a mind that involves the ability to reason,
learn, plan, think abstractly, comprehend ideas and language, and make decisions.
 It is effectively perceiving, interpreting and responding to the environment.
It is the ability to acquire and apply knowledge.
 It is the capability of observing, learning, remembering and reasoning.
 It can be defined as the ability for solving problems.
❖ Intelligence is composed of:
 Reasoning: the logical process of drawing conclusions, making predictions towards
a particular thought with the help of existing knowledge.
 Learning: a process that improves the knowledge of an AI program by making
observations about its environment.
 Problem Solving: used to searching or finding the best solution of the problems.
 Perception: the ability of machines to use input data from sensors to learn about
many aspects of the world.
Linguistic Intelligence: applications make it possible to produce, communicate and
understand any digital speech and text content on any media for everyone.
 What is Artificial Intelligence

 AI attempts to develop intelligent agents.

 Learn from experience and adopt to changing environment
 Use vast amount of knowledge
 Characteristics of Intelligent system
 Interact with human using language and speech
 Respond in real time
 Tolerate error and ambiguity in communication
❖ Therefore, AI is all about creating Intelligent agent as human intelligence.
❖ Different people approach AI with different goals in mind.
❖ Two important questions to ask are:
 Are you concerned with thinking, or behavior?
 Do you want to model humans, or try to achieve the optimal results?
❖ Definition AI is the branch of Computer Science that deals:
✓ An intelligent entity created by humans.
✓ Capable of performing tasks intelligently without being explicitly instructed.
✓ Capable of thinking and acting rationally and humanely
 What is Artificial Intelligence
❖ It is a branch of computer science concerned with
building intelligent machines which can behave like a
human, think like humans, and able to make decisions.
❖ Artificial Intelligence allows machines to model, and
even improve upon, the capabilities of the human mind.
❖ Generally, AI involves borrowing characteristics from
human intelligence, and applying them as algorithms in a
computer friendly way.
❖ AI requires a foundation of specialized hardware and
software for writing and training machine learning
algorithms.
❖ No one programming language is synonymous with
Artificial Intelligence , but a few, including Python and
Java, are the most popular.
Objectives Artificial intelligence
❖ Some of the objective of Artificial Intelligence include:
 To create intelligent expert systems that exhibit with
the capability to learn, demonstrate, explain and advice
its users.
To find solutions for complex mathematical and
scientific problems.
To shapes the future life of humans using technology
like chatbot software for customer service operations.
 To replicate human intelligence & solve knowledge-
intensive tasks efficiently.
To creates strong synergy between humans and AI so
that they can work together to enhance the capabilities
of both.
 Goals of Artificial intelligence
▪ Following are the main goals of Artificial Intelligence:
1. Replicate human intelligence
2. Solve Knowledge-intensive tasks
3. An intelligent connection of perception and action
4. Building a machine which can perform tasks that
requires human intelligence such as:
Proving a theorem
 Playing chess
Plan some surgical operation
Driving a car in traffic
5. Creating some system which can exhibit intelligent
behavior, learn new things by itself, demonstrate, explain,
and can advise to its user.
 Types of Artificial intelligence
❖ Different AI entities are built for different purposes, and that’s how they vary.
❖ AI can be classified to Type 1 and Type 2.
Artificial Intelligence type-1: Based on Capabilities
 Artificial Narrow Intelligence (ANI)
Artificial General Intelligence (AGI)
 Artificial Super Intelligence (ASI)
 Artificial Intelligence type-2: Based on functionality
 Reactive Machines
 Limited Memory
Theory of Mind
Self-Awareness
❖ AI type-1: Based on Capabilities
I. Artificial Weak or Narrow Intelligence (ANI).
 It is a goal-oriented AI trained to perform dedicated task with intelligence. .
 The most common and currently available AI is Narrow AI.
Narrow AI can fail in unpredictable ways if it goes beyond its limits.
Example of narrow AI Application
✓ Playing chess,
✓ Self-driving cars (autonomous cars)
✓ Speech & image recognition
✓ Apple Siri
AI type-1: Based on Capabilities
II. Artificial General Intelligence (AGI)
❖ General AI perform any intellectual task with efficiency like a human.
❖ It’s defined as AI which has a human-level of cognitive function,
across a wide variety of domains such as language processing, image
processing, computational functioning and reasoning and so on.
❖ The idea behind the general AI to make such a system that could be
smarter and think like a human on its own.
❖ Currently, there is no such system exists which could come under
general AI.
❖ The worldwide researchers are now focused on developing
machines with General AI.
III. Artificial Supper Intelligence (ASI)
❖ It is an outcome of general AI.
❖ Super AI is still a hypothetical concept of Artificial Intelligence.
❖ ASI surpasses human intelligence and can perform any task better
than a human.
AI type-II: Based on functionality
I. Reactive Machines
Purely reactive machines are the most basic types of Artificial
Intelligence.
Such AI systems do not store memories or past experiences for future
actions.
These machines only focus on current scenarios and react on it as per
possible best action.
Eg.
IBM's Deep Blue system
Google's AlphaGo
II. Limited Memory
 Limited memory machines can store past experiences or some data
for a short period of time.
 These machines can use stored data for a limited time period only.
Eg.
Self-driving cars
AI type-II: Based on functionality
III. Theory of Mind
 Theory of Mind AI should understand the human
emotions, people, beliefs, and be able to interact socially
like humans.
This type of AI machines are still not developed, but
researchers are making lots of efforts and improvement
for developing such AI machines.
IV. Self-Awareness
Self-awareness AI is the future of Artificial Intelligence.
 These machines will be super intelligent, and will have
their own consciousness, sentiments, and self-awareness.
 These machines will be smarter than human mind.
 Self-Awareness AI does not exist in reality still and it is a
hypothetical concept.
 Approaches to AI – Making
computer
❖ Many researchers are still trying to understand the process behind ‘how we think’.
❖ Implementing AI may need us to be aware of Philosophy, Mathematics,, Neuroscience,
Psychology, Computer Engineering and Linguistics disciples.
❖ Where is the intelligence coming from?
 The Robot Response:
A computer is void of senses and therefore symbols are meaningless to it, but a robot
with sensors can tie its symbols to its senses and thus understand symbols.
 The Brain Simulator Response:
 If we program a computer to mimic the brain (e.g., with a neural network) then the
computer will have the same ability to understand as like a human brain.
❖ In AI, we compare the brain and the computer.
❖AI is found on the premise that: –
workings of human mind can be explained in terms of computation, and
computers can do the right thing given correct premises and reasoning rules.
❖ There are four main approaches to Artificial Intelligence
Thinking Humanly
Acting Humanly
Thinking Rationally
Acting Rationally
 Thinking Humanly: Cognitive
Modeling Approach
❖ Reasons like humans do
 Programs that behave like humans
❖ Requires understanding of the internal activities of the brain
see how humans behave in certain situations and see if you
could make computers behave in that same way.
Eg.
 Write a program that plays chess.
❖ Systems that think like humans:
To say that a program thinks like a human, we must know how
humans think.
❖ We can learn about human thought in three ways:
 introspection—trying to catch our own thoughts as they go by;
 psychological experiments—observing a person in action;
brain imaging—observing the brain in action.
 Acting Humanly: The Turing Test
approach
❖ Systems that act like humans.
❖ Intelligent behavior: to achieve human-level performance in all cognitive
tasks.
❖ To pass the total Turing Test, the computer will have the following features:
NLP- to communicate successfully in a human language;
 Knowledge representation - to store what it knows or hears;
 Automated reasoning - to answer questions and to draw new conclusions;
 Machine learning to adapt to new circumstances & to detect & extrapolate
patterns.
❖ Turing viewed the physical simulation of a person as unnecessary to
demonstrate intelligence.
❖ However, other researchers have proposed additional features to pass a total
Turing Test,
 Computer Vision to perceive objects, and
 Robotics manipulate objects and move about.
 Think Rationally: The laws of
thought approach
❖ Systems that think rationally.
❖ The laws of thought approach
The study of the computations that make it
possible to perceive, reason, and act.
 It is a large list of logical statements that govern
the operation of our mind.
 The same laws can be codified and applied to
artificial intelligence algorithms.
It is a logical process and it concludes based on
symbolic logic.
 A system is rational if it thinks/does the right
thing through correct reasoning.
 Act rationally - The rational agent
approach
❖ Systems that act rationally
❖ The rational agent approach
 Rational agent acts to achieve high value and brings the best possible outcome for
any given task.
 Rational behavior: doing the right thing which is expected to maximize goal
achievement.
It is more general than using logic only: Because LOGIC + Domain knowledge
It allows extension of the approach with more scientific methodologies
❖Rational behavior: doing the right thing.
❖ The right thing: that which is expected to maximize goal achievement, given the
available information
❖ An agent is an entity that perceives and acts.
❖ A rational agent is one that acts so as to achieve the best outcome or, when there is
uncertainty, the best expected outcome
❖ A rational agent acts to achieve the best possible outcome in its present circumstances.
❖ It means that it’s a much more dynamic and adaptable agent.
❖ Doing the right thing so as to achieve one’s goal, given one’s beliefs.
❖ Rational action requires the ability to represent knowledge and reason with it so as to
reach good decision
 The Foundations of AI
❖ Intelligence is an intangible part of our brain
which is a combination of :
 Philosophy
 Mathematics
 Economics
 Neuroscience
 Psychology
 Computer engineering
 Control theory and cybernetics
 Linguistics
Applications of Artificial intelligence
❖ Artificial intelligence is the simulation of human intelligence processes by machines, especially
computer systems.
❖ The concepts of Artificial Intelligence are implemented in various fields.
❖ Specific applications of AI include:
 Speech recognition,
 Natural language processing,
 Expert systems,
 Computer vision and,
 Computer Game and Robotics.
1. Speech Recognition
❖ Speech recognition is the process of converting spoken language into text.
❖ Intelligent systems are capable of hearing and understanding the language what people are saying
in terms of sentences and their meanings.
❖ The technology uses machine learning and neural networks to process audio data and convert it
into words that can be used in businesses.
❖ It can handle different accents, slang words, noise, change in human’s noise due to cold, etc.
Applications of Artificial intelligence
2. Natural Language Processing
❖ The computer can now understand natural languages and hence human can now interact
using natural spoken languages.
❖ The ability of a computer program to understand human language as it is spoken and written
document is referred to as natural language.
❖ Some of the main functions that NLP algorithms perform are:
Text classification: Example sentiment Analysis
Machine translation: Converting from one language to other language.
Natural language generation: analyze unstructured data and automatically produce content
based meaningful data.
3. Computer Vision
❖ It is a field of AI that train computers to interpret and understand visual world.
❖ It enables systems to derive meaningful information from digital images, videos and other
visual inputs, and make recommendations based on that information.
❖ If AI enables computers to think, computer vision enables them to see, observe and
understand.
❖ Two essential technologies are used to accomplish this: a type of machine learning called
deep learning and a convolutional neural network (CNN).
❖ Example: Image classification and Object detection
Applications of Artificial intelligence
4. Expert System
❖ It is a computer program that is designed to solve complex problems and to provide
decision-making ability like a human expert.
❖ It performs this by extracting knowledge from its knowledge base using the reasoning
and inference rules according to the user queries.
 Telecommunications, where they are used to make decisions about network
technologies used and maintenance of existing networks.
Healthcare: used where they assist with medical diagnoses.
 Agriculture: used for forecast crop damage.
5. Computer Game
❖ AI in gaming refers to responsive and adaptive video game experiences.
❖ AI plays crucial role in strategic games such as chess, poker, tic-tac-toe, etc., where
machine can think of large number of possible positions based on heuristic knowledge.
❖ These are some of the best artificial intelligence games:
 Chess & Video game
 PlayStation
 The Last of Us
 Half life
 Exercise 1
1. What is AI?& what is the different between Artificial &
intelligent?
2. Intelligent is composed of
a,_______
b,_______
c,_______
d,_______
e,_______
3. List the Characteristics of Intelligent system
4. Explain the objective of Artificial Intelligence
5. What are the main goals of Artificial Intelligence?
6. List the main approaches to Artificial Intelligence & explain it
7. How many types of AI ?& explain it
8. List & explain the Applications of Artificial intelligence
Chapter 2
Intelligent Agent
 Introduction
❖Artificial intelligence is defined as the
study of rational agents.
❖A rational agent could be anything that
makes decisions, as a person, firm, machine,
or software.
❖The agents act in their environment.
❖The environment may contain other
agents.
 Agent and Environment
❖An AI system is composed of an agent and its environment.
What is Agent?
❖ An agent is anything that can be viewed as :
Perceiving its environment through sensors and,
 Acting upon that environment through effectors to maximize progress
towards its goal.
❖ Agent is task specific & specialized, well-defined Goal and Environment.
❖ It is all about decisions/ make decisions based on percept.
An agent has to choose what action to perform.
An agent has to decide when to perform an action.
 A percept is an agent’s perceptual inputs.
A percept sequence is the history of everything the agent ever received.
The agent is assumed to exist in an environment in which it perceives
and acts.
 Agent and Environment
Examples of Agent
❖ A human agent has eyes, ears, nose, tongue and other organs for sensors and hands, legs, vocal tract, and so
on for actuators.
❖ A robotic agent might have cameras and infrared range finders for sensors and various motors for actuators.
❖ A software agent receives file contents, network packets, and human input as sensory inputs and acts on the
environment
❑Environment
❖ Some programs operate in an entirely artificial environment confined to keyboard input, database, computer
file systems and character output on a screen.
❖ Examples of such environments are :
 Computer game environments,
 The internet and the
Physical world in case of robots and humans.
Task Environment
❖ PEAS : Performance measure, Environment, Actuators, and Sensors are all these under the heading of the
task environment.
❖Example: the task of designing a self-driving car PEAS is
 Performance measure: Safe, fast, comfortable trip
 Environment: Roads, other traffic, weather, pedestrians, customer.
 Actuators: Steering wheel, accelerator, brake, signal, horn
 Sensors: Cameras, LIDAR (light/radar), speedometer, GPS, odometer engine sensors, keyboard,
accelerometer, sonar.
 Properties of Task Environments
❖ In AI, Environment:
❖ Is everything which surrounds the agent, but it is not a part of an
agent itself.
❖ The environment is where agent lives, operate and provide the agent
with something to sense and act upon it.
❖ The agent takes input from the environment through sensors and
delivers the output to the environment through actuators.
❖ Can be described as a situation in which an agent is present
❖ There are six based types of environments in Artificial Intelligence:
 Fully Observable vs Partially Observable
Deterministic vs Stochastic
Single-agent vs Multi-agent
 Static vs Dynamic
Discrete vs Continuous
Episodic vs Sequential
 Properties of Task Environments
I. Observable Environment
❖ If an agent’s sensors give it access to the complete state of
the environment needed to choose an action, then the task
environment is fully observable.
❖ If the sensors detect all aspects that are relevant to the choice
of action, so an agent can determine the state of the system at all
times, then the task environment is fully observable.
❖ An environment might be Partially observable because of noisy
and inaccurate sensors or because parts of the state are simply
missing from the sensor data.
 If not, parts of the environment are inaccessible.
 Agent must make informed guesses about world.
 E.g. Vacuum Agent
❖ If the agent has no sensors at all then the environment is
unobservable environment.
 Properties of Task Environments
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.
 Deterministic environments are the environments where the next state is
observable at a given time.
 So there is no uncertainty in the environment.
 Eg:
The traffic signal
 The Stochastic environment is the opposite of a deterministic environment.
 If the environment is partially observable, then it could appear to be
stochastic
 The next state is totally unpredictable for the agent. So randomness exists
in the environment.
 Eg:
Taxi driving
The radio station
 Properties of Task Environments
III. Episodic vs Sequential
 Episodic is an environment where each state is independent of each other.
In Episodic environment, the agent’s experience is divided into atomic episodes.
 In each episode the agent receives a percept and then performs a single action.
❖ The next episode does not depend on the actions taken in previous episodes.
 Eg :
classifying images
 In Sequential environment, the current decision could affect future decisions.
 The sequential environment is an environment where the next state is
dependent on the current action.
So agent current action can change all of the future states of the environment.
 Eg :
 Playing tennis
Chess and taxi driving
 Properties of Task Environments
IV. Static vs Dynamic
 Static environments: are easy to deal with because the agent need not keep
looking at the work while it is deciding on an action, nor need it worry about the
passage of time.
 The Static environment is completely unchanged while an agent is thinking/
precepting the environment.
 Eg:
Cleaning a room (Environment) by a dry-cleaner reboot (Agent )
Crossword puzzles
 If the environment can change while an agent is deliberating, then it is dynamic
Environment for that agent.
 Dynamic Env’t could be changed while an agent is percepting the environment.
 So agents keep looking at the environment while taking action.
 Eg:
Playing soccer
Taxi driving
Chess – is semi dynamic
 Properties of Task Environments
V. Discrete vs Continuous
Discrete Environment consists of a finite number of states and agents have a
finite number of actions.
 Eg:
Choices of a move (action) in a tic-tac game
 Continuous environment, the environment can have an infinite number of states.
So the possibilities of taking an action are also infinite.
 Eg:
 basketball game
VI. Single Agent vs Multi-Agent
 Single agent environment where an environment is explored by a single agent.
All actions are performed by a single agent in the environment.
 Eg:
Playing tennis
 If two or more agents are taking actions in the environment, it is known as a
multi-agent environment.
 Eg:
 Playing a soccer
 Intelligent Agent
What is Intelligent Agent?
An intelligent agent is a goal-directed agent.
It perceives its environment through its sensors using the observations
and built-in knowledge, acts upon the environment through its actuators.
It may also be referred to as a bot, which is short for robot.
It is a program that can make decisions or perform a service based on
its environment, user input and experiences.
An autonomous entity which act upon an environment using sensors
and effectors for achieving goals.
 Discrete agent receives percepts one at a time from the environment,
and maps this percept sequence to a sequence of discrete actions .
An Agent runs in the cycle of perceiving, thinking, & acting.
 Agents interact with environments through Sensors & Effectors.
In agents that employ artificial intelligence,
 User input is collected using Sensor like microphone or cameras, and
 Agent output is delivered through Actuators, like speakers or screen.
 Intelligent Agent
 Eg:
Automated taxi driving system
 Percepts: Video, speedometer, odometer, engine
sensors, keyboard input, microphone, GPS, …
 Actions: Steer, accelerate, brake, horn, speak/
display, …
 Goals: Maintain safety, reach destination,
maximize profits (fuel, tire wear), obey laws,
provide passenger comfort, …
Environment: U.S. urban streets, freeways, traffic,
pedestrians, weather, customers, …
 Vacuum-cleaner Agent
 Acting of Intelligent Agents
(Rationality)
Rationality: Acting of Intelligent Agents.
Good Behavior: The Concept of Rationality
A rational agent is one that does the right thing.
 Means every entry in the table for agent function is correct.
Consider the consequences ..
 Agent precepts → action sequence → environmental changes
The right thing = the rational Action.
The action that maximizes the expected value of the
performance measure given the percept sequence to date.
 It is the status of being reasonable, sensible, and having good
sense of judgment.
Rationality is concerned with expected actions and results
depending upon what the agent has perceived.
 Acting of Intelligent Agents
(Rationality)
❖A rational agent is one that does the right thing.
o Measured using a metric :
1. Performance measures
2. Rationality
3. Omniscience, learning, and autonomy
Rational Agent: Performance measures
 Agent generates a sequence of actions according to the percepts it
receives.
 If the sequence is desirable, then the agent has performed well.
Generally PM: should be designed based on what is wanted in the envt
instead of how you think the agent should act.
Rational Agent: Rational 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 environments.
 The actions that the Agents can perform.
The Agent’s percept Sequences up to now (Date).
 Acting of Intelligent Agents
(Rationality)
Rational Agent: Omniscience, learning, and autonomy
 Agent can’t know in advance what the outcomes will be.
An omniscient agent knows the actual outcome of its actions & can act
accordingly.
 But omniscience (perfection) is impossible in reality.
This example shows that rationality is not the same as perfection.
Rationality is distinct from omniscience (all knowing).
 We can behave rationally even when faced with incomplete information.
Percepts may not supply all relevant information
 Rational is distinct from being perfect.
 Rationality maximizes expected outcome while omniscience (perfection)
maximizes actual outcome.
The rationality does not require omniscience, then, because the rational
choice depends only on the percept sequence to date.
An agent is autonomous if its behavior is determined by its own experience
(with ability to learn and adapt) rather than knowledge of designer.
 Acting of Intelligent Agents
(Rationality)
Generally, in rationality, An agent should
strive to "do the right thing", based on what:
 It can perceive and
 the actions it can perform.
 Rational agent should select an action (right
action) expected to maximize its
performance measure, given the evidence
provided by the percept sequence and
whatever built-in knowledge the agent has.
The right action is the one that will cause the
agent to be most successful.
 Structure of Intelligent Agents
The job of AI is to design an agent program that implements
the agent function the mapping from percepts to actions.
The agent programs take the current percept as input from the
sensors and return an action to the actuators.
The structure of an intelligent agent is a combination of
architecture and agent program.
 Agent = Architecture + Agent Program
 Architecture: is machinery that an AI agent executes on.
 Agent programs:
It is an implementation or execution of the agent function.
 Agent function: is a function in which actions are mapped
from a certain percept sequence.
 Percept sequence Complete history of everything that the
agent has ever perceived.
Percept: Agent’s perceptual inputs at any given instant.
 Structure of Intelligent Agents
Actuators: are component of machines that converts energy into motion.
The actuators are only responsible for moving and controlling a system.
 An actuator can be an electric motor, gears, rails, etc.
Sensor: a device which detects the change in the environment and sends
the information to other electronic devices.
 An agent observes its environment through sensors.
Effectors: a devices which affect the environment.
Effectors: legs, wheels, arms, fingers, wings, fins, and display screen.
 Based on their capability range and extent of intelligence, Agent programs
can be classified into five types.
 Agent programs Types
I. Simple reflex agents
II. Model based agents
III. Goal-based agents
IV.Utility-based agents
V. Learning agents
 Structure of Intelligent Agents
I. Simple reflex agents
 It is the most basic of the intelligent agents out there.
Agents do not have memory of past world states or percepts.
It selects actions & take decisions depend only on current
percept & & ignore percept history.
These agents work only if the environment is fully observable.
 It works on Condition-action rule, in which it maps the current
state to action.
 It uses just condition-action rules
o The rules are like the form “if … then …”
The agent program for a simple reflex agent in the two-
location vacuum environment.
 Structure of Intelligent Agents
 Problems for the simple reflex agent design approach.
 They have very limited intelligence
 Action does not depend on percept history, only on current percept
 They do not have knowledge of non-perceptual parts of the current state
 Mostly too big to generate and to store.
Not adaptive to changes in the environment.
II. Model-based reflex agent
 It is one that uses its percept history and its internal memory to make
decisions about an internal ''model'' of the world around it.
 It can work in a partially observable environment, and track the situation.
 A model-based agent has two important factors:
 Model: It is knowledge about "how things happen in the world,"
Internal State: It is a representation of the current state based on percept
history.
These agents have the model, "which is knowledge of the world" and based
on the model they perform actions.
 Structure of Intelligent Agents
III.Goal-Based Agents
Knowing something about the current state of the environment is not
always enough to decide what to do.
 Goal-based agent is an artificial intelligence agent that responds to its
environment and adjusts accordingly to achieve a goal.
Goal-based agent operates based on a goal in front of it and makes
decisions based on how best to reach that goal.
A model-based & goal-based agent keeps track of the world state as well as
a set of goals it is trying to achieve, and chooses an action that will
(eventually) lead to the achievement of its goals
Goal-based agents expand the capabilities of the model-based agent by
having the "goal" information.
 Such considerations of different scenario are called searching and planning,
which makes an agent proactive.
 They choose an action, so that they can achieve the goal.
 Actions chosen goals, based on
 The current state
 The current percept
 Structure of Intelligent Agents
 Uses knowledge about a goal to guide its actions: E.g., Search,
planning
 It is more flexible than reflex agents which may involve search
& planning
IV. Utility-Based Agents
 It is similar to the goal-based agent but provide an extra
component of utility measurement which makes them
different by providing a measure of success at a given state.
 It act based not only goals but also the best way to achieve
the goal.
It is useful when there are multiple possible alternatives, and
an agent has to choose in order to perform the best action.
The utility function maps each state to a real number to check
how efficiently each action achieves the goals.
 Structure of Intelligent Agents
 Utility function U: State → R indicating a measure of success
or happiness when at a given state.
Use decision theoretic models: e.g., faster vs. safer.
V. Learning Agents
 A learning agent in AI is the type of agent which can learn
from its past experiences, or it has learning capabilities.
 Learning agents: Adapt and improve over time.
 It starts to act with basic knowledge and then able to act and
adapt automatically through learning.
 More complicated when agent needs to learn utility
information:
Reinforcement learning (based on action payoff)
 Therefore, learning agents are able to learn, analyze
performance, and look for new ways to improve the
performance.
 Structure of Intelligent Agents
❖ A learning agent has mainly four conceptual
components, which are:
 Learning element: It is responsible for making
improvements by learning from environment.
 Critic: Learning element takes feedback from critic
which describes that how well the agent is doing
with respect to a fixed performance standard.
 Performance element: It is responsible for selecting
external action.
 Problem generator: This is responsible for
suggesting actions that will lead to new and
informative experiences.
Summary of Intelligent Agents
 An agent perceives and acts in an environment, has an architecture, and is
implemented by an agent program.
A rational agent always chooses the action which maximizes its expected
performance, given its percept sequence so far.
 An autonomous agent uses its own experience rather than built-in knowledge of the
environment by the designer.
An agent program maps from percept to action and updates its internal state.
Perception: the process of acquiring, interpreting, selecting, and organizing sensory
information.
 Simple Reflex agents are based on condition-action rules, respond immediately to
percepts.
 They are stateless devices which do not have memory of past world states.
 Model-based agents have internal state, which is used to keep track of past states of
the world
Goal-based agents act in order to achieve their goals, possible sequence of steps.
 Agents of this kind take future events into consideration.
 Utility-based agents maximize their own utility function.
 Learning agents have the ability to improve performance through learning.
 Representing knowledge is important for successful agent design.
 Exercise 2
1. Artificial intelligence is defined as the study of_________
2. The agents act in their_____
3. The environment may contain other_____
4. An AI system is composed of an ____and its________
5. List an examples of Agent & List an examples of env’t
6. PEAS stands:____, ______, ______,_______
7. What is different between Fully Observable & Partially Observable ?
8. What is different between Deterministic & Stochastic ?
9. What is different between Single-agent & Multi-agent ?
10.What is different between Static & Dynamic ?
11.What is different between Discrete & Continuous ?
12.What is different between Episodic & Sequential ?
13.What is different between Rationality &Omniscience?
14.The structure of an intelligent agent is a combination of _____and_____
15.What is different between sensor, actuator &effector ?
16.List & explain the Agent programs Types
 READ MORE
thanks

GOOD LUCK!

4/23/2023 Compiled by Yared Ab 48