Artificial Intelligence

5/24/2018 Artificial Intelligence

1/10

BY: LALIT KUMARCLASS: VI-E

ROLL NO.: 20

5/24/2018 Artificial Intelligence

2/10

Artificial intelligence(AI) is the human-like intelligenceexhibited by machines orsoftware. It is also an academic field ofstudy.

The central problems (or goals) of AI researchinclude reasoning, knowledge,planning, learning, natural languageprocessing(communication),perceptionand

the ability to move and manipulate objects.

5/24/2018 Artificial Intelligence

3/10

Currently popular approachesinclude statistical methods, computationalintelligenceand traditional symbolic AI. Thereare a large number of tools used in AI,

including versions of search andmathematical optimization, logic, methodsbased on probability and economics, andmany others.

5/24/2018 Artificial Intelligence

4/10

The field of AI research was founded at aconferenceon the campus of DartmouthCollegein the summer of 1956. Theattendees, includingJohn McCarthy, Marvin

Minsky, Allen Newelland Herbert Simon,became the leaders of AI research for manydecades.

They and their students wrote programs that

were, to most people, simplyastonishing: computers were solving wordproblems in algebra, proving logical theoremsand speaking English.

5/24/2018 Artificial Intelligence

5/10

Deduction, reasoning, problemsolving Early AI researchers developed

algorithms that imitated thestep-by-step reasoning thathumans use when they solvepuzzles or make logical

deductions. Knowledge representation Knowledge

representationand are centralto AI research. Many of theproblems machines areexpected to solve will requireextensive knowledge about theworld. Among the things thatAI needs to represent are:objects, properties, categoriesand relations betweenobjects;situations, events,states and time;causes andeffects; knowledge about

knowledge ; and many other,less well researched domains.

5/24/2018 Artificial Intelligence

6/10

Planning Intelligent agents must be able to set goals

and achieve them.They need a way tovisualize the future (they must have arepresentation of the state of the world andbe able to make predictions about how their

actions will change it) and be able to makechoices that maximize the utility(or "value")of the available choices.

Learning Machine learning is the study of computer

algorithms that improve automaticallythrough experience and has been central toAI research since the field's inception.

5/24/2018 Artificial Intelligence

7/10

Natural language processing communication) Natural language processinggives machines

the ability to read and understandthelanguages that humans speak. Perception Machine perceptionis the ability to use input

from sensors such as cameras,microphones, tactile sensors, sonar andothers more exotic) to deduce aspects of theworld. Computer visionis the ability toanalyze visual input. A few selectedsubproblems are speech recognition, facialrecognitionand object recognition.

5/24/2018 Artificial Intelligence

8/10

Search and optimization Many problems in AI can be

solved in theory by intelligentlysearching through manypossiblesolutions: Reasoningcan bereduced to performing asearch. For example, logicalproof can be viewed assearching for apath that leadsfrom premisesto conclusions,where each step is theapplication of an inferencerule.

Planningalgorithms searchthrough trees of goals andsubgoals, attempting to find apath to a target goal, a processcalled means-ends analysis.

5/24/2018 Artificial Intelligence

9/10

Logic is used for knowledge representation and

problem solving, but it can be applied to other

problems as well. For example,the satplanalgorithm uses logic for planningandinductive logic programmingis a methodfor learning.

Types

Default logics, non-monotoniclogicsand circumscriptionare forms of logicdesigned to help with default reasoning andthe qualification problem. Several extensions oflogic have been designed to handle specific

domains of knowledge, such as:descriptionlogics; situation calculus, eventcalculusand fluent calculus(for representingevents and time);causal calculus; belief calculus;and modal logics.

5/24/2018 Artificial Intelligence

10/10

Artificial intelligencehas been used in a wide range of fieldsincluding medical diagnosis,stock trading, robotcontrol, law, remote sensing, scientific discovery and toys.

List of applications Typical problems to which AI methods are applied Pattern recognition

Optical character recognition Handwriting recognition Speech recognition Face recognition

Artificial Creativity Computer vision, Virtual realityand Image processing Diagnosis artificial intelligence) Game theoryand Strategic planning Game artificial intelligenceand Computer game bot Natural language processing, Translationand Chatterbots Nonlinear controland Robotics