Tuesday, 1 February 2022
UNIT I: Artificial Intelligence
Definition of AI, Early work in AI, Importance of AI and related fields, Task domains of AI systems, Intelligent Agent, Generic architecture of intelligent agent.
Questions
Q. 1. Write and explain various AI problem characteristics with examples.
Q. 2. What are the issues taken into consideration in the design of search programs? Suggest any optimized solution for it.
Q. 3. What do you mean by control strategies? What are the requirements of good control strategies?
Q. 4. What is a production system? Give classification of the production system. Explain it with examples.
Q. 5. Explain the term AI and write the importance of AI. What are the task domains of the AI system?
Q. 6. Why do we need a production system? Write different types of production systems and give one example of each.
Q.7. Analyze the following problem concerning all problem characteristics.
(i) Travelling salesman problem
(ii) Block world problem
Q. 8. What are intelligent agents? Draw and explain the generic architecture of intelligent agents
Definitions of Artificial Intelligence
AI is a branch of computer science concerned with the study and creation of computer systems that exhibit some form of intelligence
It is the study of how to make computers do the things which at present humans can do better.
AI is intelligence exhibited by machines or software.
John McCarty defines AI as the “Science and engineering of making intelligent machines”
AI is accomplished by studying how human brains think and how human brains decide and work while trying to solve a problem and then writing the outcomes of this study as a basis for developing intelligent software and system
The intellectual characteristic of humans, such as the ability to reason, discover meaning, generalize, or learn from past experience.
Components of intelligence: learning, reasoning, problem-solving, perception, and using language.
Goals Of AI:
Replicate human behavior
Solve knowledge-intensive task
An intelligent connection of perception and action
Building a machine that can perform tasks that requires human intelligence such as
Theorem proving
Playing chess
Plan some surgical operations
Driving a car in traffic
Creating some system that can exhibit intelligent behavior, can learn new things by itself, demonstrate, explain, and can advise its user.
What is intelligence composed of???
Importance of AI
Error reduction
Difficult Exploration
Daily Application
Digital Assistants
Repetitive jobs
Medical Applications
No Breaks
WHAT ARE EXAMPLES OF ARTIFICIAL INTELLIGENCE?
Siri, Alexa, and other smart assistants
Self-driving cars
Robo-advisors
Conversational bots
Email spam filters
Netflix's recommendations
Classifications Of AI or Task domains of AI
Domains of AI are classified into Mundane tasks, formal tasks, and Expert tasks.
According to task domains, AI applications are divided into three major categories:
# General application
# Formal application
# Expert application
General application: - This includes perception. Perception involves sight, sound, smell & touch. Action includes the ability to understand and manipulate objects e.g., to build a robot for particular work we require to understand the process of that work.
Formal application: - Another important application of AI is to solve the formal computer application that requires intelligence and decision making during their operation, these applications are called the formal applications of AI. E.g. game playing like chess, backgammon, etc. where we have so many rules and various game positions.to win the game we require a program that can understand the game starting position, various win, and loss conditions & make decisions for every step according to the opposite player.
Another example of a formal task is to solve the complex mathematics expression of integral calculus, geometry, and theorem proving by using various lemmas and formulae.
Expert Application: - many practical applications require only specified expertise without the assistance of common-sense knowledge. These applications are called expert applications. Expert system has proven to be effective in a no of problem domain which normally requires the kind of intelligence possessed by a human expert. Expert application domains include law, chemistry, biology, operations finance, banking, metrology, geophysics, and more.
AI Problem characteristics:
There are many methods to solve AI problems but each method has limited capability. Choosing the most appropriate method is difficult & requires the implementation of multiple techniques to solve the problem. Each problem is analyzed with the following key attributes:
Is the problem decomposable to the smaller or easier problems?
Solution steps can be undone or ignored?
Is the universe predictable?
Is a good solution absolute or relative?
Is the knowledge base consistent?
Requires lots of knowledge?
Requires periodic interactions between humans and computers?
Example:1) Consider the following mathematical expression
Example:2)Consider the following facts:
i) John was a man
ii) John was a Russian.
iii) John was born in 60 A.D.
iv) All men are mortal
v) All Russian died when the volcano erupted in 90 A.D.
vi)No mortal lives more than 125 years
vii)it is now 2008 A.D.
Suppose the problem is to find out the answer “Is John alive”
Sr. No. Facts Justification
1 John was a man Fact i
2 All men are mortal iv
3 John is Mortal I & iv
4 John was born in 60 A.D. iii
5 it is now 2008 A.D vii
6 Johns age is 1948 years iii & vii
7 No mortal lives more than 125 years vi
8 John is dead 3, 6 & 8
Example: 3) Find out the shortest route that visits each city exactly once. Suppose the cities to be visited and the distance between them is as follows: -
Issues in the design of search programs:
Define the problem precisely: - This definition must include precise specification of what initial situations will be as well as what final situations constitute acceptable solutions to the problem.
Analyze the problem: - A few very important features can have an immense impact on the appropriateness of various possible techniques for solving the problem.
Isolate and represent the task knowledge that is necessary to solve the problem.
Choose the best problem-solving techniques and apply them to the particular problem.
Production System and its type
Production systems provide an excellent tool for structuring AI programs.
Production systems are highly modular because the individual rules can be added, removed, or modified independently.
The production rules of the production system are expressed in natural form.
Production systems provide a structure for describing and performing the search processes.
Def:
A production system consists of:
A set of rules, each consisting of the left side (a pattern) that determines the applicability of the rule and the right side that describes the operation to be performed if the rule is applied.
One or more knowledge/databases that contain whatever information is appropriate for the particular task. Some parts of the database may be permanent, while other parts of it may pertain only t the solution of the current problem.
A control strategy that specifies the order in which the rules will be compared to the database and a way of resolving conflicts that arise when several rules match at once.
A rule applier
Types of the Production system
Monotonic production system:
The production system in which the application of a rule never prevents the latter application of another rule that could have been applied at the first time rule was selected.
Nonmonotonic:
The production system in which the application of one rule prevents the latter application of another rule that could have been applied the first time the rule was selected.
Partial commutative:
This system states that if the application of a particular rule transforms state x into y then permutation of those rule also transform state x to y.
Commutative:
Production system which is both monotonic and partial commutative.
For any problem several production systems exist, some will be more efficient than others.
Control Strategy:
In AI problems during the process of searching for a solution to a problem, the question arises how to decide which rule to apply next. The question arrives when more than one rule will have its left side match the current state.
The first requirement of a good control strategy is that it causes motion: -
Consider the water jug problem, suppose we implement the simple control strategy of starting each time at the top of the list of rules and choosing the first applicable one. If we did that, we would never solve the problem. We would continue indefinitely filling the 4-gallon jug with water. A control strategy that does not cause motion will never lead to a solution.
The second requirement of a good control strategy is that it be systematic: -
Here is another simple control strategy for to water jug problem. On each cycle, choose at random from among the applicable rules. This strategy is better than the first, it causes motion, it will lead to a solution eventually. But we are likely to arrive at the same state a second time during the process and to use more steps than are actually necessary. Because the control strategy is not systematic. We may explore a useless sequence of operations many times before arriving at the final solution.
Examples: there are two methods to achieve the control strategies for searching
Breadth-First Search
Breadth-first is a simple strategy in which the root node
is expanded first, then all the successors of the root node are expanded next, then their successor, and so on. In general, all the nodes are expanded on a given depth in the search tree before any nodes at the next level are expanded.
Breadth-first search can be implemented by queue structure, where first in first out property is maintained means the nodes that are visited first will be expanded first
Depth First Search
Depth-first search is performed by dividing downwards into a tree. It does this by generating a child node from the most recently expanded node, then generating that child’s children, and so on until a goal node is found or some out-of-depth point is reached.
If a goal is not found i.e., when a leaf is reached or at the cutoff point, the program backtrack to the most recently expanded nodes and generates another of its children. This process continues until a goal is found or failure occurs.
Depth-first search strategy can be implemented with stack, where last in first out (LIFO)is maintained. Depth-first places the newly generated children at the head of the stack so that they will be chosen first.
Intelligent Agents
An agent is anything that can perceive its environment through sensors and acts upon that environment through effectors.
A human agent has sensory organs such as eyes, ears, nose, tongue, and skin parallel to the sensors, and other organs such as hands, legs, mouth, for effectors.
A robotic agent replaces cameras and infrared range finders for the sensors, and various motors and actuators for effectors.
A software agent has encoded bit strings as its programs and actions.
Agent Terminology
Performance Measure of Agent − It is the criteria, which determines how successful an agent is.
The behavior of Agent − It is the action that the agent performs after any given sequence of percepts.
Percept − It is the agent’s perceptual inputs at a given instance.
Percept Sequence − It is the history of all that an agent has perceived to date.
Agent Function − It is a map from the precept sequence to an action.
Rationality
Rationality is nothing but the status of being reasonable, sensible, and having a good sense of judgment.
Rationality is concerned with expected actions and results depending upon what the agent has perceived. Performing actions to obtain useful information is an important part of rationality.
What is Ideal Rational Agent?
An ideal rational agent is the one, which is capable of doing expected actions to maximize its performance measure, based on −
Its percept sequence
Its built-in knowledge base
The rationality of an agent depends on the following −
The performance measures, determine the degree of success.
Agent’s Percept Sequence till now.
The agent’s prior knowledge about the environment.
The actions that the agent can carry out.
A rational agent always performs the right action, where the right action means the action that causes the agent to be most successful in the given percept sequence. The problem the agent solves is characterized by Performance Measures, Environment, Actuators, and Sensors (PEAS).
The Structure of Intelligent Agents
Agent’s structure can be viewed as −
Agent = Architecture + Agent Program
Architecture = the machinery that an agent executes on.
Agent Program = an implementation of an agent function.
Simple Reflex Agents
They choose actions only based on the current percept.
They are rational only if a correct decision is made only based on current precepts.
Their environment is completely observable.
Condition-Action Rule − It is a rule that maps a state (condition) to an action.
Model-Based Reflex Agents
They use a model of the world to choose their actions. They maintain an internal state.
Model − knowledge about “how the things happen in the world”.
Internal State − It is a representation of unobserved aspects of the current state depending on percept history.
Updating the state requires the information about −
How the world evolves.
How the agent’s actions affect the world.
Goal-Based Agents
They choose their actions to achieve goals. The goal-based approach is more flexible than a reflex agent since the knowledge supporting a decision is explicitly modeled, thereby allowing for modifications.
Goal − It is the description of desirable situations.
Utility-Based Agents
They choose actions based on a preference (utility) for each state.
Goals are inadequate when −
There are conflicting goals, out of which only a few can be achieved.
Goals have some uncertainty of being achieved and you need to weigh the likelihood of success against the importance of a goal.
The Nature of Environments
Some programs operate in an entirely artificial environment confined to keyboard input, database, computer file systems, and character output on a screen.
In contrast, some software agents (software robots or softbots) exist in rich, unlimited softbots domains. The simulator has a very detailed, complex environment. The software agent needs to choose from a long array of actions in real-time. A softbot designed to scan the online preferences of the customer and show interesting items to the customer works in a real as well as an artificial environment.
The most famous artificial environment is the Turing Test environment, in which one real and other artificial agent are tested on equal ground. This is a very challenging environment as it is highly difficult for a software agent to perform as well as a human.
Turing Test
The success of the intelligent behavior of a system can be measured with Turing Test.
Two persons and a machine to be evaluated participate in the test. Out of the two persons, one plays the role of the tester. Each of them sits in a different room. The tester is unaware of who is a machine and who is a human. He interrogates the questions by typing and sending them to both bits of intelligence, to which he receives typed responses.
This test aims at fooling the tester. If the tester fails to determine the machine’s response from the human response, then the machine is said to be intelligent.
Comments
Post a Comment