LectureS2:ArtificialIntelligence
LectureS2:ArtificialIntelligence
3
Overview
AwhirlwindtourofArtificialIntelligence.
Wespendjustonelecture,butthereare:
EntirecourseonAI.COS302
NewAIProfessoratPrinceton.RobSchapire
Today'slevelofaspiration.
Aquicksurveyofseveralimportanttopics.
4
Origins
Ideaofprogrammingcomputersfor"intelligent"behavior.
FirstsuggestedbyAlanTuring,1950.
Term"artificialintelligence"coinedbyJohnMcCarthyin1955.
Dartmouthsummerconference,1956.
Gatheringoffield'sluminaries.
Veryoptimistic!
Optimisticpredictionsverycommonin1950sand1960s.
Actualprogressmuchslowerthanpredicted.
Somestrikingsuccesses;stilllotstodo.
J.McCarthy,1951
"Everyaspectoflearningoranyotherfeatureof
intelligencecaninprinciplebesoprecisely
describedthatamachinecanbemadetosimulateit."
5
8-puzzle
Slidetilesuntilthey'reinnumericalorder.
Ofcoursethereis:8-puzzleontheWeb.
Whatstrategytousetosolvepuzzle?
Noobviousalgorithm.
Fromthisposition,
Threesuccessorpositions.
Fromeachofthese,
–
Two,three,orfoursuccessors.
–
Andsoon.
Soundslikeatree?
Yup.
6
2
8
4
1
3
5
7
6
8-puzzle:Breadth-FirstSearch
SearchtreeforGoalnode,onelevelatatime.
7
8-puzzle:Depth-FirstSearch
SearchforGoaldowntosomedepth(thesearch"horizon").
8
8-puzzle:HeuristicSearch
Breadth-FirstandDepth-Firstare"blind"searches.
Exhaustivemethodsforfinding
pathtogoal.
Ofteninfeasiblebecausetoomany
nodesexpanded.
Success(eventually)guaranteed.
"Heuristic"search.
Uses"evaluationfunction"torank
successornodes;pickbest.
Noguaranteeofsuccess.
Exampleusesdistancefromstart
plusnumberoftilesoutofplace.
Manyotherfunctionspossible.
Note:only6nodesexpanded.
9
GameTrees
Whatifyouhaveanopponent?
"Minimax"methods.
Youchoosebestmoveassumingopponentchoosestheirbestresponse.
Opponentchoosestheirbestmoveassumingyoushouldbestresponse.
Youchooseyour
bestmove.(max)
Opponentchooses
bestreply.(min)
Youchoose...
-6
24
5
-3
8
2
-6
-3
2
2
10
Tic-Tac-Toe
Partialgametree.
11
Tic-Tac-Toe
Reducetreebyrecognizingsymmetries.
12
SizesofGameTrees
8-puzzle.
Firsttilecanbeinanyoneof9squares.
Secondtileinoneof8.
Totalnumberofnodes=9!/2=181,440
Tic-tac-toe(ignoringsymmetries).
Firstmove:oneof9spots.
Secondmove:oneof8.
Somegamesterminatebefore9movesmade.
Totalnumberofnodes<9!=362,880
Bothnumberssmall,soexhaustivesearchfeasible.
Butwhataboutsomebiggergame,forinstance...
14
Chess
AfavoritetargetofAIresearcherssince1950's.
Howbigisgametree?
20possiblefirstmoves.
35possiblenextmovesonaverage.
–
called"branchingfactor"
Supposegamesaverage40moves(eachside).
Completegametreethereforehas:
–
3580
≥
≥≥
≥10120
nodes!
Soifeachelectronintheuniversewereasupercomputer,etc.,etc.
Anycomputer(andanyperson)cansearchonlythetiniestfractionof
thecompletegametree.
15
Chess
Lotsofeffort,andsteadyimprovements,since1950's.
DeepBluechessmachinedevelopedatIBM.
Hsu,Campbell,Tan,Hoane,Brody,Benjamin,1990's.
32-processorparallelcomputer.
512specialchesschips.
Evaluates200millionpositions/sec.
Hugeamountofchessknowledgeprogrammedin.
Usescleverheuristicfunctionsandminimaxtreesearch.
GarryKasparovvs.DeepBlue(1997)
VladimirKramnikvs.DeepFritz(2002)
GarryKasparovvs.DeepJunior(2003)
16
Backgammon
Backgammon.
Movesuserollof2dice.
–
21differentoutcomes
Manypossiblemovesperdiceroll.
–
about20onaverage
Branchingfactorofabout400.
–
toobig!
Searchesofevenmoderatedepthcomputationallyprohibitive.
GoodheuristicfunctionVERYimportant.
Somethingbettermustbedone...
17
Uses
neuralnetwork
and
reinforcementlearning.
G.Tesauro,1995.
Eachconnectionhasaweight.
Eachnodeoutputsweightedsumofinputs.
Weightsadjustedateachstep.
Inputpatterns.(198)
Describeboardposition.
Outputpatterns.(1)
Probabilityofwinningfromgiven
boardposition.
Weightadjustmentfunction.
Triestomakesuccessivepredictionsagree.
Hence:temporaldifference.
TD-Gammon
22
LoebnerContestDialogs
"Hello,Judge.MynameisElla.Pleasetellmeyourfirstname,or
whateverIshouldcallyou."
"CallmeLisa.AmImaleorfemale?"
"Okay,Iwillcallyou'Lisa.AmIMaleOrFemale?'Iamhererecently
fromTainjin,China,sopleasebepatientwithme."
"I'mnotapatientbecauseIamnotill.Areyoufeelingwelltoday?
"Whatdoyouthinkitmeansthatyouarenotill?"
"Haveyoueverbeenill?"
"'Haveyoueverbeenill?'yousay?That'sit,LisaAmIMaleOr
Female!Iwillscheduleawhippingforyou."
23
LoebnerContestDialogs
"WhatdoyouthinkofReading?"
"I'msomewhatcrazyaboutreading.Iliketoreadmysterynovels."
"Imeantthetown."
"Joneslivesinthecityandisawriterofchildren'sbooks."
"DoesJonesliveinReading,then?"
"Tomorrowisanotherday.Wecanonlylivetodaywell."
"Doyoulive,then?"
"Ilivethenalot.Somedaywewillliveonanotherplanet."
"Ithinkyouarealreadythere."
24
LoebnerContestDialogs
"Whatisit?"
"Whatiswhat?"
"Ahigh-yield,perennial,four-lobedhybridofwheatandryewhose
rootgrain,triticale,cantraceitsancestrytotwentieth-century
Canada."
"Atlast.Anotherclassicgrainfreak."
"Lewis.JimLewis.Wouldyoulikeformetoshowyousomepoetry?"
"No,thanks."
"Ididn'twanttoshowyouthestupidpoetryanyway."
25
ChineseRoomExperiment
Imaginethatyou:
Don'tunderstandChinese.
Arealoneinaroomthathaspaperslotslabeledinputandoutput.
HaveabigbookofChinesewriting.
HaveEnglishinstructions(notranslations)thattellyouwhatto
writeonyouroutputpaperinresponsetovariousinputs.
Andthen:
Chinesespeakersoutsidetheroompassinpiecesofpaperwith
Chinesewriting.Theyknowthesearequestions(butyoudon't).
Youconsultyourmanualofinstructions,figureouttheproper
Chineseresponse,copyitdown,andpassitout.
TheBigQuestion.Searle,1980.
ThefolksoutsidethinkyouunderstandChinese.Doyou?
Ifacomputerdidthesame,woulditunderstandChinese?
26
ChineseRoomExperiment
"Weak"AIhypothesis.
MachinescanbeprogrammedtoEXHIBITintelligentbehavior.
Surelytrue:DeepBlue,TD-Gammon,others.
Programsusemethodsverydifferentfromhumans.
–
performance(oftask)vs.simulation(ofhumanmethods).
"Strong"AIhypothesis.
MachinescanbeprogrammedtoPOSSESSintelligence.
Musttheyusebrain-likemethods(e.g.,neuralnets)?
–
"Connectionism"
SearleusedChineseRoomasabsoluterefutationofthepossibilityof
strongAI.
Butmanydisagree!
27
"Reverse"TuringTest
StandardTuringTest:judgeishuman.
ReverseTuringTest:judgeiscomputer!
Why?
Yahooallowseachuser15MbytesofWebstorage.
–
Youwritea"bot"totosignup1millionusers.
–
Congratulations.Younowhave15Terabytesofstorage!
PayPal
offers$5foreachuserwhoopensanewaccount.
–
Youwriteabottosignup1billionusers.
–
Congratulations.Younowhave$5,000,000,000!
Onlinepolls.
Spamfiltering.
Allneedtodistinguishrealhumansfrombots(programs).
How?captcha.net
28
AIQuotes
"JustastheWrightbrothersatKittyHawkin1903wereontheright
tracktothe747,sotooisAI,withitsattemptstoformalize
commonsenseunderstanding,onitswaytofullyintelligent
machines."PatrickWinston
"Believingthatwritingthesetypesofprogramswillbringuscloserto
realartificialintelligenceislikebelievingthatsomeoneclimbingatree
ismakingprogresstowardreachingthemoon."HubertDreyfus
"Thebrainhappenstobeameatmachine."MarvinMinsky
"Eitherartificialintelligence
is
possible...or
we're
not."HerbSimon
"AIisanythinginsoftwarethatwedon'tknowhowtodoyet."
"Thequestionofwhetheracomputercanthinkisnomoreinteresting
thanthequestionofwhetherasubmarinecanswim."E.W.Dijkstra
29
AIinEverydayLife
ManyexamplesofAImethodsatworkintherealworld.
MicrosoftOffice'shelpfultalkingpaperclip.
R.I.P.?
Speechrecognition.
Speakslowlyandclearlytothetelephonerobot.
Opticalcharacterrecognition(OCR).
MakesU.S.PostalServicehappy.
Controlofspacecraft!
AIsystemgivencontrolof
DeepSpace1for2daysinMay1999.
30
OmittedTopics
Knowledgerepresentation.
Reasoning.
Expertsystems.
Naturallanguageunderstanding.
Speechrecognition.
Computervision.
Anddozensmore...
(Buthey,weonlyhadtheonelecture.)
Enter the password to open this PDF file:
File name:
-
File size:
-
Title:
-
Author:
-
Subject:
-
Keywords:
-
Creation Date:
-
Modification Date:
-
Creator:
-
PDF Producer:
-
PDF Version:
-
Page Count:
-
Preparing document for printing…
0%
Σχόλια 0
Συνδεθείτε για να κοινοποιήσετε σχόλιο