Notes on Introduction to Artificial Intelligence - Slides | CS 440, Study notes of Computer Science

Material Type: Notes; Professor: Draper; Class: Introduction to Artificial Intelligence; Subject: Computer Science; University: Colorado State University; Term: Fall 2008;

Typology: Study notes

Pre 2010

Uploaded on 03/19/2009

koofers-user-on3-1
koofers-user-on3-1 🇺🇸

10 documents

1 / 4

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
1
1
1
CS440
CS440 -
-Introduction to
Introduction to
Artificial Intelligence
Artificial Intelligence
Lecture #1
Lecture #1
8/26/08
8/26/08
2
2
Administrivia
Administrivia
Textbook: S. Russell and
Textbook: S. Russell and
P.
P. Norvig
Norvig.
. Artificial
Artificial
Intelligence: A Modern
Intelligence: A Modern
Approach
Approach. Prentice Hall,
. Prentice Hall,
2003, 2nd edition. ISBN:
2003, 2nd edition. ISBN:
0
0-
-13
13-
-790395
790395-
-2
2
Get to know this book. It will be your friend.
3
3
Administrivia
Administrivia (II)
(II)
Web Site:
Web Site:
The class web site is
The class web site is
www.cs.colostate.edu/~cs440
www.cs.colostate.edu/~cs440
All lectures posted there (after presentation)
All lectures posted there (after presentation)
All homework assignments posted there
All homework assignments posted there
Important
Important
news
news
bulletins
bulletins
Anonymous feedback page
Anonymous feedback page
4
4
Administrivia
Administrivia (III)
(III)
Contact information
Contact information
Instructor: Bruce A. Draper
Instructor: Bruce A. Draper
Office: 236 USC
Office: 236 USC
This will change to 442 CSB
This will change to 442 CSB
Office Hours: TBD
Office Hours: TBD
Phone: 491
Phone: 491-
-7873 (hard to reach me this way)
7873 (hard to reach me this way)
Email:
[email protected] (much easier)
(much easier)
Teaching Assistant: Jason Remington
Teaching Assistant: Jason Remington
Email:
Email: jason.remington.ta@gmail.com
5
5
Administrivia
Administrivia (IV)
(IV)
Workload
Workload
Two Exams: one midterm, one final.
Two Exams: one midterm, one final.
Final is Tuesday, Dec. 16
Final is Tuesday, Dec. 16th
th, 11:20am
, 11:20am-
-1:20pm
1:20pm
Three programming assignments
Three programming assignments
All solo
All solo
Two (
Two (re)written
re)written assignments
assignments
Graded both on content and style
Graded both on content and style
Each written assignment has three parts
Each written assignment has three parts
An initial written assignment
An initial written assignment
A critique of a colleague
A critique of a colleague
s written assignment
s written assignment
A rewritten assignment
A rewritten assignment
All three parts will be graded
All three parts will be graded
6
6
Administrivia
Administrivia (V)
(V)
Grading
Grading
Final: 25%
Final: 25%
Midterm: 15%
Midterm: 15%
Programming Assignments: 40%
Programming Assignments: 40%
Written Assignments: 20%
Written Assignments: 20%
Curving
Curving
Yes, grades are curved
Yes, grades are curved
During the semester you get numeric scores and
During the semester you get numeric scores and
class histograms; curve is applied at the end
class histograms; curve is applied at the end
pf3
pf4

Partial preview of the text

Download Notes on Introduction to Artificial Intelligence - Slides | CS 440 and more Study notes Computer Science in PDF only on Docsity!

11

CS440 - CS440- Introduction toIntroduction to

Artificial IntelligenceArtificial Intelligence

Lecture #1Lecture

22

Administrivia Administrivia

Textbook: S. Russell and Textbook: S. Russell and

P. NorvigP.Norvig.. ArtificialArtificial

Intelligence: A ModernIntelligence: A Modern

ApproachApproach. Prentice Hall,. Prentice Hall,

2003, 2nd edition. ISBN:2003, 2nd edition. ISBN:

Get to know this book. It will be your friend.

33

AdministriviaAdministrivia (II)(II)

Web Site:Web Site:

  • – The class web site isThe class web site is www.cs.colostate.edu/~cs440www.cs.colostate.edu/~cs
  • – All lectures posted there (after presentation)All lectures posted there (after presentation)
  • – All homework assignments posted thereAll homework assignments posted there
  • – ImportantImportant ““newsnews”” bulletinsbulletins
  • – Anonymous feedback pageAnonymous feedback page

44

Administrivia (III) Administrivia(III)

Contact information Contact information

  • – Instructor: Bruce A. DraperInstructor: Bruce A. Draper Office: 236 USCOffice: 236 USC - This will change to 442 CSB–This will change to 442 CSB Office Hours: TBDOffice Hours: TBD Phone: 491Phone: 491--7873 (hard to reach me this way)7873 (hard to reach me this way) Email:Email: [email protected]@cs.colostate.edu (much easier)(much easier)
  • – Teaching Assistant: Jason RemingtonTeaching Assistant: Jason Remington Email:Email: [email protected]@gmail.com

55

Administrivia (IV) Administrivia(IV)

WorkloadWorkload

  • – Two Exams: one midterm, one final.Two Exams: one midterm, one final. Final is Tuesday, Dec. 16Final is Tuesday, Dec. 16 thth^ , 11:20am-, 11:20am-1:20pm1:20pm
  • – Three programming assignmentsThree programming assignments All soloAll solo
  • – Two (Two (re)writtenre)written assignmentsassignments Graded both on content and styleGraded both on content and style Each written assignment has three partsEach written assignment has three parts - – An initial written assignmentAn initial written assignment - – A critique of a colleagueA critique of a colleague’’s written assignments written assignment - – A rewritten assignmentA rewritten assignment All three parts will be gradedAll three parts will be graded

66

Administrivia (V) Administrivia(V)

Grading Grading

  • – Final: 25%Final: 25%
  • – Midterm: 15%Midterm: 15%
  • – Programming Assignments: 40%Programming Assignments: 40%
  • – Written Assignments: 20%Written Assignments: 20%

Curving Curving

  • – Yes, grades are curvedYes, grades are curved
  • – During the semester you get numeric scores andDuring the semester you get numeric scores and class histograms; curve is applied at the endclass histograms; curve is applied at the end

77

Administrivia (VI) Administrivia(VI) Cheating PolicyCheating Policy

  • – It is your responsibility to know the department cheating policyIt is your responsibility to know the department cheating policy
  • – Yes, we will use MOSS and/or other tools to help us detect cheatYes, we will use MOSS and/or other tools to help us detect cheatinging
  • –––^ When in doubt, ask meWhen in doubt, ask meVery rare at this level, but taken very seriously when it happenVery rare at this level, but taken very seriously when it happenss OnOn--time policytime policy
  • – Deadlines are deadlines (no extensions)Deadlines are deadlines (no extensions)
  • – PrePre--extensions may be granted for good reasonsextensions may be granted for good reasons
  • –^ ““UnforseeableUnforseeable emergenciesemergencies”” happenhappen –– see mesee me Cell phones : off or mutedCell phones : off or muted
  • –^ If muted, leave the room before answeringIf muted, leave the room before answering Asking questions in classAsking questions in class
  • ––– Strongly encouragedStrongly encouragedPlease give your name first (so I can put names to faces)Please give your name first (so I can put names to faces)
  • – I may choose to take some discussions offI may choose to take some discussions off--lineline

88

About This Course About This Course

Divided roughly into three sections Divided roughly into three sections

  • – SearchSearch
  • – LogicLogic
  • – Bayesian InferenceBayesian Inference

Machine learning gets its own follow- Machine learning gets its own follow-onon

course (CS545)course (CS545)

99

For Thursday For Thursday

Read Chapter 1 of your textRead Chapter 1 of your text

  • – It should remind you of todayIt should remind you of today’’s discussion.s discussion.
  • – It will be useful for the 1It will be useful for the 1^ stst^ written assignment.written assignment.

Read Chapter 2 of your textRead Chapter 2 of your text

  • – Background for ThursdayBackground for Thursday’’s lectures lecture

1010

What is AI?What is AI?

“The exciting new effort to make “The exciting new effort to make

computers think …computers think… machines with minds,machines with minds,

in the full and literal sense”in the full and literal sense” Haugeland,Haugeland,

“(The automation of) activities that we “(The automation of) activities that we

associate with human thinking, activitiesassociate with human thinking, activities

such as decision making, problem solving,such as decision making, problem solving,

learning …learning…..”” Bellman, 1978Bellman, 1978

How different are these?

1111

What is AI? (II) What is AI? (II)

The definitions vary by:The definitions vary by:

  • – Thought processes vs. actionThought processes vs. action
  • – Judged according to human standards vs. success accordingJudged according to human standards vs. success according to an ideal concept of intelligence: rationality.to an ideal concept of intelligence: rationality.

Systems that actSystems thatact rationallyrationally

Systems that actSystems thatact likelike humanshumans

Systems that thinkSystems thatthink rationallyrationally

Systems that thinkSystems thatthink likelike humanshumans

Definitions of artificial intelligence:

1212

AI: Act Like Humans AI: Act Like Humans

“The art of creating machines that perform “The art of creating machines that perform

functions that require intelligence whenfunctions that require intelligence when

performed by people”performed by people” Kurzweil, 1990Kurzweil, 1990

“The study of how to make computers do “The study of how to make computers do

things which, at the moment, people arethings which, at the moment, people are

better at”better at” Rich and Knight, 1991Rich and Knight, 1991

1919

AI in everyday lifeAI in everyday life

AI systems we use on a daily basis:AI systems we use on a daily basis:

  • – Scanning documents (OCR).Scanning documents (OCR).
  • – Talking with customer service or yourTalking with customer service or your cellphonecellphone (voice recognition, NLP).(voice recognition, NLP).
  • – Applying for a loan.Applying for a loan.
  • – Online map services.Online map services.
  • – Computer games (chess etc.)Computer games (chess etc.)
  • – Web searchesWeb searches

2020

AI Systems AI Systems

Deep Blue Deep Blue

  • – Defeats Kasparov, Chess Grand MasterDefeats Kasparov, Chess Grand Master -- IBM 1997IBM 1997 Deep Space 1 Deep Space 1
  • – AI planner controls space probeAI planner controls space probe -- NASA 1999.NASA 1999. DARPA grand challenge 2005: a 130 mile race DARPA grand challenge 2005: a 130 mile race of driverless cars in the mojaveof driverless cars in themojave desert.desert.

http://www.roadtripamerica.com/GettingOutThere/DARPA-Grand-Challenge-2005-2.htm

2121

Mundane vs. Expert Tasks Mundane vs. Expert Tasks

MundaneMundane

  • – Identifying objects in a visual imageIdentifying objects in a visual image
  • – Answering a questionAnswering a question
  • – Picking up an eggPicking up an egg

ExpertExpert

  • – ChessChess
  • – Medical diagnosisMedical diagnosis
  • – Configuring computer hardware (circuitConfiguring computer hardware (circuit layout)layout) 2222

Foundations of AI Foundations of AI

Philosophy: Philosophy: Logic, reasoning, rationality.Logic, reasoning, rationality. Mathematics: Mathematics: Logic, computability, tractabilityLogic, computability, tractability Psychology: Psychology: understanding how humans think and act.understanding how humans think and act. Neuroscience: Neuroscience: (^) how do brains process information?how do brains process information? Economics: Economics: theory of rational decisions, game theorytheory of rational decisions, game theory .. Computer Engineering: Computer Engineering: building the hardware andbuilding the hardware and software that make AIsoftware that make AI Linguistics: Linguistics: how to deal with languagehow to deal with language … …

2323

CSU AI Faculty CSU AI Faculty

Machine LearningMachine Learning (Bioinformatics)(Bioinformatics)

AsaAsa BenBen--HurHur

Computer VisionComputer Vision ((BiomimeticsBiomimetics))

Bruce DraperBruce Draper

Computer VisionComputer Vision (Face recognition)(Face recognition)

RossRoss BeveridgeBeveridge

Adele HoweAdele Howe Agents & PlanningAgents & Planning

Machine LearningMachine Learning (Markov Models)(Markov Models)

Chuck AndersonChuck Anderson

Machine LearningMachine Learning (Genetic Algorithms)(Genetic Algorithms)

Darrell WhitleyDarrell Whitley