【MOOC】人工智能原理-北京大學(xué) 中國大學(xué)慕課MOOC答案_第1頁
【MOOC】人工智能原理-北京大學(xué) 中國大學(xué)慕課MOOC答案_第2頁
【MOOC】人工智能原理-北京大學(xué) 中國大學(xué)慕課MOOC答案_第3頁
【MOOC】人工智能原理-北京大學(xué) 中國大學(xué)慕課MOOC答案_第4頁
【MOOC】人工智能原理-北京大學(xué) 中國大學(xué)慕課MOOC答案_第5頁
已閱讀5頁,還剩18頁未讀 繼續(xù)免費(fèi)閱讀

下載本文檔

版權(quán)說明:本文檔由用戶提供并上傳,收益歸屬內(nèi)容提供方,若內(nèi)容存在侵權(quán),請(qǐng)進(jìn)行舉報(bào)或認(rèn)領(lǐng)

文檔簡介

【MOOC】人工智能原理-北京大學(xué)中國大學(xué)慕課MOOC答案QuizzesforChapter11、【單選題】TuringTestisdesignedtoprovidewhatkindofsatisfactoryoperationaldefinition?圖靈測試旨在給予哪一種令人滿意的操作定義本題答案:【machineintelligence機(jī)器智能】2、【多選題】Selectthefollowingtruestatementsregardingtheconceptofartificialintelligence選擇以下關(guān)于人工智能概念的正確表述本題答案:【Artificialintelligenceistheintelligenceexhibitedbymachinesorsoftware.人工智能是通過機(jī)器或軟件展現(xiàn)的智能#Artificialintelligenceaimstocreateintelligentmachines.人工智能旨在創(chuàng)造智能機(jī)器#Artificialintelligenceistoenablethedevelopmentofcomputersthatareabletodothingsnormallydonebyhuman.人工智能是為了開發(fā)一類計(jì)算機(jī)使之能夠完成通常由人類所能做的事#Artificialintelligenceistostudyandconstructagentprogramsthatperformwellinagivenenvironment.人工智能是研究和構(gòu)建在給定環(huán)境下表現(xiàn)良好的智能體程序】3、【多選題】Whichofthefollowingarethefoundationsofartificialintelligence?如下學(xué)科哪些是人工智能的基礎(chǔ)?本題答案:【Philosophy哲學(xué)#Mathematics數(shù)學(xué)#Economics經(jīng)濟(jì)學(xué)#Psychology心理學(xué)】4、【多選題】WhichofthefollowingstatementsaretruetodescribestrongAI(GeneralAI)?下列陳述中哪些是描述強(qiáng)AI(通用AI)的正確答案?本題答案:【Theappropriatelyprogrammedcomputerwiththerightinputsandoutputswouldtherebyhaveamindinexactlythesamesensehumanbeingshaveminds.是經(jīng)過適當(dāng)編程的具有正確輸入和輸出的計(jì)算機(jī),因此有與人類同樣判斷力的頭腦#Itmeansamachinewiththeabilitytoapplyintelligencetoanyproblem.指的是一種機(jī)器,具有將智能應(yīng)用于任何問題的能力】5、【多選題】Selectthefollowingcomputersystemswhichareinstancesofartificialintelligence.選擇下列計(jì)算機(jī)系統(tǒng)中屬于人工智能的實(shí)例本題答案:【Robot機(jī)器人#Intelligentpersonalassistant智能個(gè)人助理#WebsearchenginesWeb搜索引擎】6、【多選題】Selectwhichofthefollowingonesareresearchareaofartificialintelligence.選擇下列哪些是人工智能的研究領(lǐng)域本題答案:【Expertsystem專家系統(tǒng)#Machinelearning機(jī)器學(xué)習(xí)#Imageunderstanding圖像理解#Facerecognition人臉識(shí)別】7、【多選題】Examinesomeapplicationsofartificialintelligence(AI)todiscoverwhichthefollowingtaskscancurrentlybesolvedbyAI.考察人工智能(AI)的一些應(yīng)用,去發(fā)現(xiàn)目前下列哪些任務(wù)可以通過AI來解決本題答案:【Playingadecentgameoftabletennis(Ping-Pong).打一場像樣的乒乓球比賽#Buyingaweek’sworthofgroceriesontheWeb.在Web上購買一周的食品雜貨#PlayingthegameofTexasHoldemPokeratacompetitivelevel.以競技水平玩德州撲克游戲#TranslatingspokenEnglishintospokenChineseinrealtime.將英文口語實(shí)時(shí)翻譯為中文口語】8、【填空題】Rationallymeansthepropertyofasystemwhichdoestherightthinggivenwhatit________.理性指的是一個(gè)系統(tǒng)的屬性,即在_________的環(huán)境下做正確的事。本題答案:【knows已知】QuizzesforChapter21、【多選題】Forthefollowingcomparisons,selectwhichapproachesaremoregeneralandmoreamenabletoscientificdevelopment.對(duì)于如下比擬,選擇哪些方法更通用并且更適合于科學(xué)開發(fā)本題答案:【Thinkingrationally理性思考#Actingrationally理性動(dòng)作】2、【多選題】Arationalagentisonethatdoestherightthing,butwhatdoesitmeantodotherightthing?Consideringtheconsequencesoftheagent’sbehavior,selectthefollowingtruestatementsregardingtheconceptof"rightthing".理性智能體指的是有正確行為的智能體,但正確的行為意味著什么?考慮智能體行為的后果,選擇下列有關(guān)“正確的行為”概念的真實(shí)陳述本題答案:【Thissequenceofactionscausestheenvironmenttogothroughasequenceofstates.這些動(dòng)作序列引起環(huán)境改變而產(chǎn)生狀態(tài)序列#Anagentinanenvironmentgeneratesasequenceofactionsaccordingtothepercepts.智能體在某個(gè)環(huán)境中依據(jù)感知生成動(dòng)作序列#Ifthesequenceisdesirable,thentheagenthasperformedwell.如果該序列是所期望的,則該智能體表現(xiàn)良好】3、【多選題】PEASisataskenvironmentdescriptionwhichstandsfor:PEAS是一種任務(wù)環(huán)境描述,它代表:H、Sensors感受器本題答案:【Performance性能#Environment環(huán)境#Actuators動(dòng)作器#Sensors感受器】4、【多選題】Whichofthefollowingstatementsaretrueaboutenvironmenttypesforintelligentagents?下列中語句哪些與智能體環(huán)境類型有關(guān)本題答案:【Fullyobservablevs.partiallyobservable完全可觀測與部分可觀測#Singleagentvs.multi-agent單智能體與多智能體#Episodicvs.sequential陣發(fā)性與連續(xù)性#Deterministicvs.stochastic確定性與隨機(jī)性】5、【多選題】Thinkingthedifferencesbetweengentfunctionsandagentprograms,selectcorrectstatementsfromfollowingones.考慮智能體函數(shù)與智能體程序的差異,從下列陳述中選擇正確的答案本題答案:【Anagentprogramimplementsanagentfunction一個(gè)智能體程序?qū)崿F(xiàn)一個(gè)智能體函數(shù)#Anagentprogramcontainsagentfunction.智能體程序包含智能體函數(shù)】6、【多選題】Selectthefollowingtruewaysthatcanbeusedtorepresentstatesforanagent.選擇下列哪些方法可以用于表示智能體的狀態(tài)本題答案:【Atomic原子式#Factored因子式#Structured結(jié)構(gòu)式】7、【填空題】Theapproachofartificialintelligence(AI)whichepresentsinformationthroughsymbolsandtheirrelationshipsis_______AI,andwhichrepresentsinformationinadistributedformwithinanetworkisConnectionistAI.憑借符號(hào)及他們之間的關(guān)系來表征信息的人工智能方法是____________AI,而用網(wǎng)絡(luò)內(nèi)部的一種分布式形式來表征信息的是聯(lián)結(jié)主義AI。本題答案:【Symbolist符號(hào)主義】8、【填空題】Anintelligentagentisanythingthatcanbeviewedasperceivingitsenvironmentthrough_______,andactinguponthatenvironmentthroughactuators.智能體可看作是通過_________感知外部環(huán)境,并且通過執(zhí)行器作用于外部環(huán)境的任何事物。本題答案:【sensors感受器】QuizzesforChapter31、【單選題】Therearetwomainkindsofformulationfor8-queensproblem.Whichofthefollowingitemsissuchaformulationthatstartswithall8queensontheboardandmovesthemaround?有兩種8皇后問題的形式化方式?!俺跏紩r(shí)8個(gè)皇后都放在棋盤上,然后再進(jìn)行移動(dòng)”是哪一種形式化方式?本題答案:【Complete-stateformulation全態(tài)形式化】2、【單選題】Leth(n)denoteheuristicfunctionandg(n)denotecost,theevaluationfunctionf(n)usedbyGreedysearchis:設(shè)h(n)表示啟發(fā)式函數(shù)且g(n)表示代價(jià),則貪婪搜索所使用的評(píng)價(jià)函數(shù)是本題答案:【f(n)=h(n)】3、【單選題】Leth(n)denoteheuristicfunctionandg(n)denotecost,theevaluationfunctionf(n)usedbyA*searchis:設(shè)h(n)表示啟發(fā)式函數(shù)且g(n)表示代價(jià),則A*搜索所使用的評(píng)價(jià)函數(shù)是:本題答案:【f(n)=g(n)+h(n)】4、【多選題】Whichofthefollowingitemsareusedtoformulateaproblem?下列哪些項(xiàng)用于對(duì)問題進(jìn)行形式化H、Agentfunction智能體函數(shù)I、Goaltest目標(biāo)檢測J、Pathcost路徑檢測本題答案:【Initialstate初始狀態(tài)#Actions動(dòng)作#Transitionmodel轉(zhuǎn)換模型#Goaltest目標(biāo)檢測#Pathcost路徑檢測】5、【多選題】Selectwhichofthefollowingsearchmethodsaretrueuninformedsearch.選擇如下哪些搜索方法屬于無信息搜索H、Uniform-costsearch一致代價(jià)搜索本題答案:【Bidirectionalsearch雙向搜索#Breadth-firstsearch寬度優(yōu)先搜索#Depth-firstsearch深度優(yōu)先搜索#Depth-limitedsearch深度受限搜索#Uniform-costsearch一致代價(jià)搜索】6、【多選題】Thesearchstrategiesthatuseproblem-specificknowledgebeyondthedefinitionoftheproblemitselfareknownas:除了問題本身的定義之外,使用問題特定知識(shí)的搜索策略被認(rèn)為是本題答案:【Informedsearch有信息搜索#Heuristicsearch啟發(fā)式搜索】7、【填空題】Some______problemsorNP-hardproblemscanbesolvedonlybysearch.某些_______或NP難問題只能通過搜索來求解。本題答案:【NP-completeNP完】8、【填空題】Problemformulationistheprocessofdecidingwhat_______andstatestoconsider,givenagoal.問題形式化是給定一個(gè)目標(biāo),決定要考慮的_______與狀態(tài)的處理。本題答案:【actions動(dòng)作】QuizzesforChapter41、【單選題】Keepingjustonenodeinmemorymightseemtobeanextremereactiontotheproblemofmemorylimitations.Localbeamsearchkeepstrackof:在內(nèi)存中僅保存一個(gè)節(jié)點(diǎn)似乎是對(duì)內(nèi)存限制問題的極端反應(yīng)。局部束搜索保持:本題答案:【kstatesratherthanjustone.k個(gè)狀態(tài)而不僅僅為一】2、【單選題】Ageneticalgorithmisavariantofstochasticbeamsearch,inwhichsuccessorstatesaregenerated:遺傳算法是隨機(jī)束搜索的一個(gè)變體,其中后繼節(jié)點(diǎn)的生成是由:本題答案:【bycombiningtwoparentstatesratherthanbymodifyingasinglestate.組合兩個(gè)雙親狀態(tài)而不是修改單一狀態(tài)。】3、【多選題】Hillclimbingsearchissometimescalledgreedylocalsearchbecauseitgrabsagoodneighborstatewithoutthinkingaheadaboutwheretogonext.Unfortunately,itoftengetsstuckforthethreereasons:爬山搜索有時(shí)也被稱為貪婪局部搜索,因?yàn)樗活欁プ∫粋€(gè)好的鄰接點(diǎn)的狀態(tài),而不提前思考下一步該去哪兒。它在三種情況下經(jīng)常被困:本題答案:【Localmaxima局部最大值#Plateaux高原#Ridges山嶺】4、【多選題】WhichofthefollowingstatementsaretrueaboutSimulatedannealingalgorithm?以下關(guān)于模擬退火算法的陳述哪些是正確的?本題答案:【Theinnerloopofthesimulatedannealingalgorithmisquitesimilartohillclimbing.模擬退火算法的內(nèi)循環(huán)與爬山法非常相似。#Insteadofpickingthebestmove,simulatedannealingalgorithmpicksarandommove.模擬退火算法不是選擇最佳行動(dòng),而是選擇隨機(jī)行動(dòng)?!?、【填空題】AntColonyOptimization(ACO)algorithmwasinspiredbythebehaviorofantsseekingapathbetween_______andasourceoffood.蟻群優(yōu)化算法是受螞蟻在_______和食物源之間尋找路徑行為的啟發(fā)而形成的。本題答案:【betweentheirnest蟻巢】6、【填空題】Inspiredbysocialbehaviorofbirdsandfishes,ParticleSwarmOptimization(PSO)algorithmusesanumberof________thatconstituteaswarmmovingaroundinthesearchspacelookingforthebestsolution.受鳥類和魚類的社會(huì)行為的啟發(fā),粒子群優(yōu)化算法采用若干_______構(gòu)成一個(gè)圍繞搜索空間移動(dòng)的群體來尋找最優(yōu)解。本題答案:【particles粒子】7、【填空題】Localsearchalgorithmsoperateusingasingle______(ratherthanmultiplepaths)andgenerallymoveonlytoneighborsofthatnode.局部搜索算法使用一個(gè)______(而不是多條路徑),并且通常僅移動(dòng)到該節(jié)點(diǎn)相鄰的節(jié)點(diǎn)。本題答案:【currentnode當(dāng)前節(jié)點(diǎn)】8、【填空題】Inadditiontofindinggoals,localsearchalgorithmsareusefulforsolvingpure_________,inwhichtheaimistofindthebeststateaccordingtoanobjectivefunction.除了尋找目標(biāo)之外,局部搜索算法對(duì)解決純_________也很有效。其目的是根據(jù)一個(gè)目標(biāo)函數(shù)找到其最好的狀態(tài)。本題答案:【optimizationproblems優(yōu)化問題】QuizzesforChapter51、【單選題】Whichofthefollowingisatruestatementaboutgames?以下關(guān)于博弈的陳述哪個(gè)是正確的?本題答案:【Adversarialsearchproblemsareoftenknownasgames.對(duì)抗搜索通常稱為博弈】2、【單選題】___________describesasituationinwhichtheinteractingagents'aggregategainsandlossescanbelessthanormorethanzero.___________中智能體交互動(dòng)作的總收益和損失可以小于或大于零本題答案:【Non-zerosumgame非零和博弈】3、【多選題】Selectthefollowingtruestatementsregardingtheconceptofminimaxruleforazerosumgame.從如下關(guān)于零和博弈maximum概念中選擇正確的答案。本題答案:【Eachplayermaximizesthemaximumpayoffpossibleforitself.每個(gè)玩家會(huì)使自己可能的最大收益變得最大。#Eachplayermaximizesthemaximumlosspossiblefortheother.每個(gè)玩家會(huì)使對(duì)手可能的最大損失變得最大?!?、【多選題】Whichofthefollowingstatementsaretrueaboutalpha-betapruning?以下關(guān)于alpha–beta剪枝的陳述哪些是正確的?本題答案:【Alpha–betapruningistodecreasethenumberofnodesthatareevaluatedbytheminimaxalgorithminitssearchtree.Alpha–beta剪枝旨在減少其搜索樹中由minimax算法評(píng)價(jià)的節(jié)點(diǎn)數(shù)量。#Alpha–betapruningistoeliminatelargepartsthatareevaluatedbytheminimaxalgorithminitssearchtree.Alpha–beta剪枝旨在消除其搜索樹中由minimax算法評(píng)價(jià)的大部分?!?、【填空題】ClaudeShannonproposedinsteadthatprogramsshouldcutoffthesearchearlierandapplya____________evaluttionfunctiontostatesinthesearch,effectivelyturningnonterminalnodesintoterminalleaves.克勞德·香農(nóng)提出:程序應(yīng)該早一些剪斷搜索,并在搜索中對(duì)狀態(tài)應(yīng)用_____________評(píng)價(jià)函數(shù),有效地將非終端節(jié)點(diǎn)轉(zhuǎn)換為終端葉節(jié)點(diǎn)。(請(qǐng)?zhí)顚懼形拇鸢福┍绢}答案:【啟發(fā)式】6、【填空題】____________isadynamicgamewithprobabilistictransitionsplayedbyoneormoreplayers.____________是一種具有概率轉(zhuǎn)換的動(dòng)態(tài)博弈,有一個(gè)或多個(gè)玩家。(請(qǐng)?zhí)顚懼形拇鸢福┍绢}答案:【隨機(jī)博弈】7、【填空題】Monte-Carlomethodsareabroadclassofcomputationalalgorithmsthatrelyon________________toobtainnumericalresults.蒙特卡羅方法是一大類計(jì)算算法,它憑借________________來獲得數(shù)值結(jié)果。(請(qǐng)?zhí)顚懼形拇鸢福┍绢}答案:【重復(fù)隨機(jī)采樣】8、【填空題】___________treesearchisontheanalysisofthemostpromisingmoves,expandingthesearchtreebasedonrandomsamplingofthesearchspace.___________樹搜索對(duì)最有利的動(dòng)作進(jìn)行分析,根據(jù)搜索空間的隨機(jī)采樣來擴(kuò)展搜索樹。(請(qǐng)?zhí)顚懼形拇鸢福┍绢}答案:【蒙特卡羅】QuizzesforChapter61、【單選題】Selectthefollowingtrueonethatisusedtothestaterepresentationforconstraintsatisfactionproblems(CSPs).從如下用于約束滿足問題(CSP)的狀態(tài)表示中選擇正確的答案。本題答案:【Factored因子】2、【單選題】Assumethat{A,B,C,D}arevariables,thedomainofeachvariableis{u,v,w},and!=denotesnotequal,selectwhichofthefollowingexpressionsisabinaryconstraintonCSPformalism?設(shè){A,B,C,D}為變量,每個(gè)變量的域是{u,v,w},且“!=”表示不等于,從如下表達(dá)式中選擇那個(gè)是CSP形式化的2元約束?本題答案:【Diff(A,D)】3、【多選題】CompareCSPandstate-spacesearch,andselectcorrectstatementsfromfollowingones.比較CSP和狀態(tài)空間搜索,并從下列敘述中選擇正確的答案。本題答案:【CSPsolvingsystemcanbefasterthanstate-spacesearchsolvingsystem.CSP求解系統(tǒng)會(huì)比狀態(tài)空間搜索求解系統(tǒng)快。#CSPcanquicklyeliminatelargeswatchesofthesearchspace.CSP可以快速排除大的搜索空間樣本。】4、【多選題】Whichofthefollowingstatementsaretruetypesoftypesoflocalconsistencyforconstraintpropagation?如下陳述中哪些是約束傳播局部一致性的正確類型?本題答案:【Arcconsistency弧一致#k-consistencyk一致#Nodeconsistency節(jié)點(diǎn)一致#Pathconsistency路徑一致】5、【多選題】SelectthefollowingtruestatementsregardingtheconceptofBacktrackingsearch.從如下有關(guān)“回溯搜索”概念中選擇正確的答案。本題答案:【Itchoosesvaluesforonevariableatatimeandbacktrackswhenavariablehasnolegalvalueslefttoassign.每次為變量選擇值并且當(dāng)變量沒有合法賦值時(shí)回溯。#Itincrementallybuildscandidatestothesolutions,andabandonseachpartialcandidatec,assoonasitdeterminesthatccannotpossiblybecompletedtoaninvalidsolution.遞增地構(gòu)建解的候選,并且一旦確定部分候選c不能成為合法的解,就將c拋棄?!?、【填空題】Aconstraintsatisfactionproblem(CSP)isdefinedasasetofobjectswhosestatemustsatisfyanumberof_______________.約束滿足問題(CSP)被定義為其狀態(tài)必須滿足若干_______________的一組對(duì)象。本題答案:【constraintsorlimitations約束和限制】7、【填空題】Inchoosinganewvalueforavariable,min-conflictsheuristicistoselectthevaluethatresultsinthe_________withothervariables.在對(duì)一個(gè)變量選擇一個(gè)新值時(shí),最小沖突啟發(fā)式選擇導(dǎo)致與其它變量呈現(xiàn)__________的值。本題答案:【minimumnumberofconflicts最少?zèng)_突】8、【填空題】Toreduceaconstraintgraphtoatreestructure,twoapproachescanbeusedthatare___________andtreedecomposition.為了簡化約束圖為樹結(jié)構(gòu),有兩種方法可以采用,即___________和樹分解。本題答案:【cutsetconditioning割集調(diào)節(jié)】QuizzesforChapter71、【單選題】Whichofthefollowinghasbeendefinedwithreferencetoinformation?下列中哪一項(xiàng)是根據(jù)信息來定義的?本題答案:【Knowledge知識(shí)】2、【單選題】Whatkindofknowledgewillbeusedtodescribehowaproblemissolved?哪種知識(shí)可用于描述問題如何求解?本題答案:【Proceduralknowledge過程性知識(shí)】3、【單選題】Selectwhichofthefollowingapproachesaretrueusingfirstorderlogic.選擇以下哪種方法可以使用一階邏輯。本題答案:【Declarativeapproaches陳述性方法】4、【多選題】Selectthefollowingtruestatementsregardingtheconceptofsemanticnetwork.選擇如下關(guān)于“語義網(wǎng)絡(luò)”概念的正確陳述。本題答案:【Itcanrepresentthesemanticrelationsbetweenconcepts.可以表示概念之間的語義關(guān)系。#Itcanberepresentedasadirectedorundirectedgraph,consistingofnodesusedtorepresentconcepts,andarcstorepresentsemanticrelationsbetweentheconcepts.可以被表示為一種直接或間接的圖,由表示概念的節(jié)點(diǎn)和表示概念之間語義關(guān)系的弧組成。#Itcanbeusedwhenonehasknowledgethatisbestunderstoodasasetofconceptsthatarerelatedtooneanother.語義網(wǎng)絡(luò)可以用于當(dāng)所具有的知識(shí)可被很好地理解為一組彼此相關(guān)的概念時(shí)。】5、【多選題】Selectwhichofthefollowingstatementsaretrueaboutontology.選擇下列關(guān)于本體的陳述哪些是正確的。本題答案:【Anontologyisaformalnaminganddefinitionofthetypes,properties,andinterrelationshipsoftheentitiesforaparticulardomainofdiscourse.本體是一種對(duì)特定論域中實(shí)體的類型、特性和相互關(guān)系的形式化命名和定義。#Anontologyprovidesacommonvocabularyofanareaanddefinethemeaningofthetermsandtherelationshipsbetweenthem.本體提供一個(gè)領(lǐng)域的公共詞匯,并且定義一些術(shù)語的含義和它們之間的關(guān)系?!?、【多選題】ThesynonymsofBayesiannetworksinclude:貝葉斯網(wǎng)絡(luò)的同義詞包括:本題答案:【Beliefnetwork信念網(wǎng)絡(luò)#Causalnetwork因果網(wǎng)絡(luò)#Probabilisticnetwork概率網(wǎng)絡(luò)】7、【填空題】A______________systemconsistsofaknowledgebaseandaninferenceengine,whereknowledgebaserepresentsfactsabouttheworld,inferenceenginecanreasonaboutthosefacts.______________系統(tǒng)由知識(shí)庫和搜索引擎組成,其中,知識(shí)庫表示關(guān)于世界的事實(shí),推理引擎則可以對(duì)這些事實(shí)進(jìn)行推理。本題答案:【knowledgebase知識(shí)庫】8、【填空題】__________representationfocusondesigningcomputerrepresentationsthatcaptureknowledgeabouttheworldthatcanbeusedtosolvecomplexproblems.__________表示關(guān)注于設(shè)計(jì)計(jì)算機(jī)表示來采集關(guān)于世界的知識(shí),可用于解決復(fù)雜的問題。本題答案:【knowledge知識(shí)】QuizzesforChapter81、【單選題】Learnedthedifferencesbetweensingle-agentproblemandmulti-agentproblem,selectacorrectoneformfollowingdescriptions.學(xué)習(xí)了單智能體問題和多智能體問題之間的區(qū)別后,從如下描述中選擇正確的答案。本題答案:【Bothofmulti-effectorproblemandmulti-bodyproblemisasingle-agentproblem.多效用器和多軀體這兩者都是單智能體問題。】2、【多選題】Selectthefollowingcorrectfeaturesregardingclassicalplanning.選擇如下有關(guān)經(jīng)典規(guī)劃的正確特征。本題答案:【Keepingadeterministicactionatatime.每次保持確定性的動(dòng)作。#Oneagentwithonlyoneknowninitialstate.一個(gè)智能體僅有一個(gè)已知的初始狀態(tài)?!?、【多選題】Whichofthefollowingapproachesareusedforclassicalplanning?下列方法中哪些可用于經(jīng)典規(guī)劃?本題答案:【Booleansatisfiability布爾可滿足性#constraintsatisfaction約束滿足#first-orderlogicaldeduction一階邏輯推理#planninggraph規(guī)劃圖#planrefinement規(guī)劃精進(jìn)#state-spacesearch狀態(tài)空間搜索】4、【多選題】Plannersusedinrealworldforplanningandschedulingaremorecomplex,sothatselectthefollowingcorrectthingswhichshouldbeextend:現(xiàn)實(shí)世界中用于規(guī)劃和調(diào)度的規(guī)劃者更為復(fù)雜,因此選擇以下哪些應(yīng)該擴(kuò)展。本題答案:【therepresentationlanguage表示語言#thewaytointeractwithenvironment與環(huán)境交互的方式】5、【多選題】Thecharacteristicsofmulti-agentare:多智能體的特征是:本題答案:【Autonomy自主性#Decentralization分散化】6、【多選題】Tosimulatetheflockingbehaviorofbirds,therulesshouldconsistof要仿真鳥群的群體行為,其規(guī)則應(yīng)包括:本題答案:【Alignment對(duì)齊#Cohesion聚集#Separation分離】7、【填空題】Therearetwoapproachestosearchingforaplan,oneis________________search,andtheotherisbackwardrelevant-statessearch.有兩種搜索計(jì)劃的方式,一個(gè)是_____________搜索,而另一個(gè)是后向狀態(tài)空間搜索。本題答案:【forwardstate-space前向狀態(tài)空間】8、【填空題】MarkovDecisionProcess(MDP)isa_______________process,meansactionoutcomesdependonlyonthecurrentstate.馬可夫決策過程(MDP)是一種_______________過程,意味著動(dòng)作結(jié)果僅僅依賴于當(dāng)前狀態(tài)。本題答案:【discretetimestochasticcontrol離散時(shí)間隨機(jī)控制】9、【填空題】TherearetwooptimalpoliciesforDynamicProgramming,oneis______________,andtheotherispolicyiteration.動(dòng)態(tài)規(guī)劃有兩種優(yōu)化策略,一個(gè)是___________,而另一種是策略迭代。本題答案:【valueiteration值迭代】QuizzesforChapter91、【單選題】Whichofthefollowingisusedtodiscovergeneralfactsfromtrainingexamples?下列中哪個(gè)用于從訓(xùn)練樣本中發(fā)現(xiàn)一般的事實(shí)?本題答案:【Inductivelearning歸納學(xué)習(xí)】2、【單選題】ThereisasetofWebpagesidentifiedasinterestedoruninterestedfortheuserthatcanbeusedtosolveWebpagesfilteringproblembymachinelearning,selectwhichofthefollowingelement/featureistheformaldefinitionforthoseWebpages.有一組供用戶標(biāo)識(shí)為感興趣或不感興趣的網(wǎng)頁,可被機(jī)器學(xué)習(xí)用來解決網(wǎng)頁過濾問題,選擇下列特征中哪個(gè)是通過訓(xùn)練這些網(wǎng)頁學(xué)來的?本題答案:【Experience(E)經(jīng)驗(yàn)】3、【單選題】Whichofthefollowingisusedforlearnedfeatures?以下哪一個(gè)是可用于從圖像中識(shí)別視覺模式的深度神經(jīng)網(wǎng)絡(luò)?H、SIFTSIFTI、SURFSURFJ、SVMSVM本題答案:【CNNCNN】4、【多選題】Selectthefollowingtruealgorithms(methods)whichhavebeeninventedinmachinelearning?選擇下列哪些是機(jī)器學(xué)習(xí)中發(fā)明的算法(方法)?本題答案:【AdaBoost自適應(yīng)提升#DecisionTree決策樹#Perceptron感知機(jī)#RandomForests隨機(jī)森林】5、【多選題】Selectthefollowingtruepartiesregardingthethreepartiesofmachinelearning.選擇如下哪些派別是有關(guān)機(jī)器學(xué)習(xí)的三個(gè)派別?本題答案:【Connectionism聯(lián)結(jié)主義#Behaviorism行為主義#Symbolicism符號(hào)主義】6、【多選題】Selectthefollowingtruephrasesregardingthetasksthathavebeensolvedbymachinelearning.選擇下列哪些屬于可以通過機(jī)器學(xué)習(xí)解決的任務(wù)。本題答案:【Classification分類#Clustering聚類#Regression回歸】7、【多選題】Selectthefollowingtruephrasesregardingthetypicalparadigmsinmachinelearning.選擇下列哪些是典型的機(jī)器學(xué)習(xí)范式。本題答案:【Reinforcementlearning強(qiáng)化學(xué)習(xí)#Supervisedlearning有監(jiān)督學(xué)習(xí)#Unsupervisedlearning無監(jiān)督學(xué)習(xí)】8、【填空題】Machinelearningisa______ofartificialintelligence,concernstheconstructionandstudyofsystemsthatcanlearnfromdata.機(jī)器學(xué)習(xí)是人工智能的一個(gè)______,從事構(gòu)建和研究可以從數(shù)據(jù)中學(xué)習(xí)的系統(tǒng)。本題答案:【branch分支】QuizzesforChapter101、【單選題】Whichstatementbestdescribesthetaskof“classification”inmachinelearning?哪一個(gè)是機(jī)器學(xué)習(xí)中“分類”任務(wù)的準(zhǔn)確描述?本題答案:【Toassignacategorytoeachitem.為每個(gè)項(xiàng)目分配一個(gè)類別?!?、【單選題】Whichstatementbestdescribesthetaskof“regression”inmachinelearning?哪一個(gè)是機(jī)器學(xué)習(xí)中“回歸”任務(wù)的準(zhǔn)確描述?本題答案:【Topredictarealvalueforeachitem.預(yù)測每個(gè)項(xiàng)目實(shí)際的值?!?、【單選題】Whichstatementbestdescribesthetaskof“ranking”inmachinelearning?哪一個(gè)是機(jī)器學(xué)習(xí)中“排名”任務(wù)的準(zhǔn)確描述?本題答案:【Toorderitemsaccordingtosomecriterion.依據(jù)某個(gè)準(zhǔn)則對(duì)項(xiàng)目進(jìn)行排序。】4、【單選題】Whichstatementbestdescribesthetaskof“clustering”inmachinelearning?哪一個(gè)是機(jī)器學(xué)習(xí)中“聚類”任務(wù)的準(zhǔn)確描述?本題答案:【Togroupdataobjects.對(duì)數(shù)據(jù)對(duì)象進(jìn)行分組?!?、【單選題】Whichstatementbestdescribesthetaskof“densityestimation”inmachinelearning?哪一個(gè)是機(jī)器學(xué)習(xí)中“密度估計(jì)”任務(wù)的準(zhǔn)確描述?本題答案:【Tofindthedistributionofinputsinsomespace.發(fā)現(xiàn)某個(gè)空間中輸入的分布。】6、【單選題】Whichstatementbestdescribesthetaskof“dimensionalityreduction”inmachinelearning?哪一個(gè)是機(jī)器學(xué)習(xí)中“降維”任務(wù)的準(zhǔn)確描述?本題答案:【Tosimplifyinputsbymappingthemintoalowerspace.將其映射到低維空間來簡化輸入?!?、【填空題】Theoutputofregressionisareal______value,andtheoutputofclassificationisadiscretecategories.回歸的輸出是一個(gè)實(shí)際的______值,而分類的輸出則是一個(gè)離散的類別。本題答案:【continuous連續(xù)】8、【填空題】Clusteringistoidentifysimilargroupsforinputobjects_____trainingdata,andclassificationistoassignpre-definedclassesforinputitemswithtrainingdata.聚類是為輸入對(duì)象標(biāo)識(shí)相似的組,_____訓(xùn)練數(shù)據(jù);而分類則是給輸入項(xiàng)分派預(yù)定義的類,有訓(xùn)練數(shù)據(jù)。本題答案:【without沒有】QuizzesforChapter111、【單選題】Whichstatementbestdescribestheparadigmof“Supervisedlearning”inmachinelearning?哪一個(gè)是機(jī)器學(xué)習(xí)中“有監(jiān)督學(xué)習(xí)”范式的準(zhǔn)確描述?本題答案:【Itreceivesasetoflabeleddataandmakespredictionsforallunseendata.接收一組標(biāo)注數(shù)據(jù)并對(duì)所有的未知數(shù)據(jù)做出預(yù)測?!?、【單選題】Whichstatementbestdescribestheparadigmof“semi-supervisedlearning”inmachinelearning?哪一個(gè)是機(jī)器學(xué)習(xí)中“半監(jiān)督學(xué)習(xí)”范式的準(zhǔn)確描述?本題答案:【Itisaclassofsupervisedlearningthatalsomakeuseofunlabeleddatafortraining.屬于有監(jiān)督學(xué)習(xí)算法一類,此外還利用未標(biāo)記數(shù)據(jù)進(jìn)行訓(xùn)練?!?、【單選題】Whichstatementbestdescribestheparadigmof“unsupervisedlearning”inmachinelearning?哪一個(gè)是機(jī)器學(xué)習(xí)中“無監(jiān)督學(xué)習(xí)”范式的準(zhǔn)確描述?本題答案:【Itreceivesexclusivelyunlabeleddataandmakespredictionsforallunseendata.僅僅接收未標(biāo)注數(shù)據(jù)并對(duì)所有的未知數(shù)據(jù)做出預(yù)測?!?、【單選題】Whichstatementbestdescribestheparadigmof“weaklysupervisedlearning”inmachinelearning?哪一個(gè)是機(jī)器學(xué)習(xí)中“弱監(jiān)督學(xué)習(xí)”范式的準(zhǔn)確描述?本題答案:【Itaimstolearnsomeinformationusingalimitedamountoftrainingexamples.旨在采用有限數(shù)量的訓(xùn)練樣本來學(xué)習(xí)一些信息?!?、【單選題】Whichstatementbestdescribestheparadigmof“one-shotlearning”inmachinelearning?哪一個(gè)是機(jī)器學(xué)習(xí)中“一次性學(xué)習(xí)”范式的準(zhǔn)確描述?本題答案:【Itaimstolearnsomeinformationfromone,oronlyafew,trainingexamples.旨在從一個(gè)、或僅有的幾個(gè)訓(xùn)練樣本中學(xué)習(xí)一些信息?!?、【單選題】Whichstatementbestdescribestheparadigmof“zero-shotlearning”inmachinelearning?哪一個(gè)是機(jī)器學(xué)習(xí)中“零次性學(xué)習(xí)”范式的準(zhǔn)確描述?本題答案:【Itisabletosolveataskdespitenothavingreceivedanytrainingexamplesofthattask.能夠求解一個(gè)任務(wù),即使沒有得到該任務(wù)的任何訓(xùn)練樣本?!?、【單選題】Whichlearningparadigmdoesitapplystoringknowledgetoadifferentbutrelatedproblem?那種學(xué)習(xí)范式把儲(chǔ)備的強(qiáng)知識(shí)用于不同但相關(guān)的問題?本題答案:【Transferlearning遷移學(xué)習(xí)】8、【多選題】Selectthefollowingtruealgorithms(methods)inwhicheachonehasbeeninventedasanalgorithm(method)ofreinforcementlearning?選擇下列哪些算法(方法)已被發(fā)明為專門用于強(qiáng)化學(xué)習(xí)的算法。本題答案:【AsynchronousAdvantageActor-Critic異步優(yōu)勢(shì)動(dòng)作者?評(píng)論者#DeterministicPolicyGradients確定性策略梯度#NeuralEpisodicControl神經(jīng)情景控制】QuizzesforChapter121、【多選題】Selectthefollowingtruewordsinwhicheachonehasbeeninventedasatypicallearningmodelformachinelearning.從如下模型中選擇已被用作機(jī)器學(xué)習(xí)的代表性模型。本題答案:【Geometricmodel幾何模型#Logicalmodel邏輯模型#Networkedmodel網(wǎng)絡(luò)化模型#Probabilisticmodel概率模型】2、【多選題】Selectthefollowingtruephrasesinwhicheachonehasbeeninventedasaprobabilisticmodelformachinelearning.從如下短語中選擇已被用作機(jī)器學(xué)習(xí)的概率模型。本題答案:【Gaussianprocesses高斯過程#HiddenMarkovmodels隱藏馬可夫模型#NaiveBayesandBayesNetwork樸素貝葉斯和貝葉斯網(wǎng)絡(luò)#Probabilisticprogram概率規(guī)劃】3、【多選題】Whichofthefollowingalgorithmsaretrueonesusedbymanifoldlearning?下列哪些是真正被流形學(xué)習(xí)所使用的算法?本題答案:【Isomap(Isometricmapping)Isomap(等距映射)#LLE(LocallyLinearEmbeddings)LLE(局部線性嵌入)#LE(LaplacianEigenmaps)LE(拉普拉斯特征映射)】4、【多選題】Selectthefollowingphrasesinwhicheachonehastrulybeeninventedorusedasalogicalmodelformachinelearning.從如下短語中選擇真正地被發(fā)明為或被用作機(jī)器學(xué)習(xí)邏輯模型的方法。本題答案:【Associationrules關(guān)聯(lián)規(guī)則#Decisiontree決策樹#First-orderlogic一階邏輯】5、【多選題】WhichofthefollowingphrasesaretheartificialneuralnetworkstrulyusedinmachineLearning?下列短語哪些是真正用于機(jī)器學(xué)習(xí)的人工神經(jīng)網(wǎng)絡(luò)?本題答案:【Convolutionalneuralnetwork卷積神經(jīng)網(wǎng)絡(luò)#Deepauto-encoder深度自動(dòng)編碼器#Longshort-termmemory長短期記憶】6、【填空題】Theprobabilisticmodelistouseprobabilitytheorytoexpressall________________.概率模型是采用概率論來表示所有________________。本題答案:【formsofuncertainty不確定性的形式】7、【填空題】Ageometricmodelisconstructeddirectlyininstancespacewhichcanbethoughtasthe________________suchasEuclideanspaceorRiemannianspace.幾何模型直接在實(shí)例空間構(gòu)建,該空間可被認(rèn)為是________________,例如:歐幾里德空間或黎曼空間。本題答案:【geometricconcepts幾何的概念】8、【填空題】Thelogicalmodelsaredefinedintermsofeasilyinterpretable______________,orcanbeeasilytranslatedintorulesthatareunderstandablebyhumans.邏輯模型被定義為易于解釋的______________,或者易于轉(zhuǎn)換成人類能夠理解的規(guī)則。本題答案:【logicalexpressions邏輯表達(dá)式】9、【填空題】Anartificialneuralnetworkisanartificialrepresentationofthe____________thattriestosimulateitslearningprocessing.人工神經(jīng)網(wǎng)絡(luò)是一種____________的人工表示,試圖仿真其學(xué)習(xí)過程。本題答案:【humanbrain人腦】FinalExamofPoAI1、【單選題】TuringTestisdesignedtoprovidewhatkindofsatisfactoryoperationaldefinition?圖靈測試旨在給予哪一種令人滿意的操作定義?本題答案:【machineintelligence機(jī)器智能】2、【單選題】Thinkingthedifferencesbetweenagentfunctionsandagentprograms,selectcorrectstatementsfromfollowingones.考慮智能體函數(shù)與智能體程序的差異,從下列陳述中選擇正確的答案。本題答案:【Anagentprogramimplementsanagentfunction.一個(gè)智能體程序?qū)崿F(xiàn)一個(gè)智能體函數(shù)?!?、【單選題】Therearetwomainkindsofformulationfor8-queensproblem.Whichofthefollowingoneistheformulationthatstartswithall8queensontheboardandmovesthemaround?有兩種8皇后問題的形式化方式?!俺跏紩r(shí)8個(gè)皇后都放在棋盤上,然后再進(jìn)行移動(dòng)”屬于哪一種形式化方式?本題答案:【Complete-stateformulation全態(tài)形式化】4、【單選題】Whatkindofknowledgewillbeusedtodescribehowaproblemissolved?哪種知識(shí)可用于描述如何求解問題?H、Noneoftheaboveiscorrect以上都不正確本題答案:【Proceduralknowledge過程性知識(shí)】5、【單選題】Whichofthefollowingisusedtodiscovergeneralfactsfromtrainingexamples?下列中哪個(gè)用于訓(xùn)練樣本中發(fā)現(xiàn)一般的事實(shí)?本題答案:【Inductivelearning歸納學(xué)習(xí)】6、【單選題】Whichstatementbestdescribesthetaskof“classification”inmachinelearning?哪一個(gè)是機(jī)器學(xué)習(xí)中“分類”任務(wù)的正確描述?本題答案:【Toassignacategorytoeachitem.為每個(gè)項(xiàng)目分配一個(gè)類別?!?、【單選題】Whichstatementbestdescribesthetaskof“regression”inmachinelearning?哪一個(gè)是機(jī)器學(xué)習(xí)中“回歸”任務(wù)的正確描述?本題答案:【Topredictarealvalueforeachitem.預(yù)測每個(gè)項(xiàng)目實(shí)際的值?!?、【單選題】Whichstatementbestdescribesthetaskof“dimensionalityreduction”inmachinelearning?哪一個(gè)是機(jī)器學(xué)習(xí)中“降維”任務(wù)的正確描述?本題答案:【Tosimplifyinputsbymappingthemintoalowerspace.將其映射到低維空間來簡化輸入?!?、【單選題】Whichstatementbestdescribestheparadigmof“supervisedlearning”inmachinelearning?哪一個(gè)是機(jī)器學(xué)習(xí)中“有監(jiān)督學(xué)習(xí)”范式的正確描述?本題答案:【Itreceivesasetoflabeleddataandmakespredictionsforallunseendata.接收一組標(biāo)注數(shù)據(jù)并對(duì)所有的未知數(shù)據(jù)做出預(yù)測?!?0、【單選題】Whichstatementbestdescribestheparadigmof“unsupervisedlearning”inmachinelearning?哪一個(gè)是機(jī)器學(xué)習(xí)中“無監(jiān)督學(xué)習(xí)”范式的正確描述?本題答案:【Itreceivesexclusivelyunlabeleddataandmakespredictionsforallunseendata.僅僅接收未標(biāo)注數(shù)據(jù)并對(duì)所有的未知數(shù)據(jù)做出預(yù)測?!?1、【單選題】Whichlearningparadigmdoesitapplystoringknowledgetoadifferentbutrelatedproblem?哪種學(xué)習(xí)范式把儲(chǔ)備的知識(shí)用于不同但相關(guān)的問題?本題答案:【Transferlearning遷移學(xué)習(xí)】12、【多選題】Selectthefollowingtruestatementsregardingtheconceptofartificialintelligence.選擇以下關(guān)于人工智能概念的正確表述。本題答案:【Artificialintelligenceistheintelligenceexhibitedbymachinesorsoftware.人工智能是通過機(jī)器或軟件展現(xiàn)的智能。#Artificialintelligenceaimstocreateintelligentmachines.人工智能旨在創(chuàng)造智能機(jī)器。#Artificialintelligenceistoenablethedevelopmentofcomputersthatareabletodothingsnormallydonebyhuman.人工智能是為了開發(fā)一類計(jì)算機(jī)使之能夠完成通常由人類所能做的事。#Artificialintelligenceistostudyandconstructagentprogramsthatperformwellinagivenenvironment.人工智能是研究和構(gòu)建在給定環(huán)境下表現(xiàn)良好的智能體程序?!?3、【多選題】WhichofthefollowingstatementsaretruetodescribestrongAI(GeneralAI)?下列陳述中哪些是描述強(qiáng)AI(通用AI)的正確答案?本題答案:【Theappropriatelyprogrammedcomputerwiththerightinputsandoutputswouldtherebyhaveamindinexactlythesamesensehumanbeingshaveminds.是經(jīng)過適當(dāng)編程的具有正確輸入和輸出的計(jì)算機(jī),因此有與人類同樣判斷力的頭腦。#Itmeansamachinewiththeabilitytoapplyintelligencetoanyproblem.指的是一種機(jī)器,具有將智能應(yīng)用于任何問題的能力?!?4、【多選題】Forthefollowingcomparisons,selectwhichapproachesaremoregeneralandmoreamenabletoscientificdevelopment.對(duì)于如下比擬,選擇哪些方法更通用并且更適合于科學(xué)開發(fā)。本題答案:【Thinkingrationally理性思考#Actingrationally理性動(dòng)作】15、【多選題】Arationalagentisonethatdoestherightthing,butwhatdoesitmeantodotherightthing?Consideringtheconsequencesoftheagent’sbehavior,selectthefollowingtruestatementsregardingtheconceptofrightthing.理性智能體指的是有正確行為的智能體,但正確的行為意味著什么?考慮智能體行為的后果,選擇下列有關(guān)“正確的行為”概念的真實(shí)陳述。本題答案:【Anagentinanenvironmentgeneratesasequenceofactionsaccordingtothepercepts.智能體在某個(gè)環(huán)境中依據(jù)感知生成動(dòng)作序列。#Thissequenceofactionscausestheenvironmenttogothroughasequenceofstates.這些動(dòng)作序列引起環(huán)境改變而產(chǎn)生狀態(tài)序列。#Ifthesequenceisdesirable,thentheagenthasperformedwell.如果該序列是所期望的,則該智能體表現(xiàn)良好?!?6、【多選題】Sel

溫馨提示

  • 1. 本站所有資源如無特殊說明,都需要本地電腦安裝OFFICE2007和PDF閱讀器。圖紙軟件為CAD,CAXA,PROE,UG,SolidWorks等.壓縮文件請(qǐng)下載最新的WinRAR軟件解壓。
  • 2. 本站的文檔不包含任何第三方提供的附件圖紙等,如果需要附件,請(qǐng)聯(lián)系上傳者。文件的所有權(quán)益歸上傳用戶所有。
  • 3. 本站RAR壓縮包中若帶圖紙,網(wǎng)頁內(nèi)容里面會(huì)有圖紙預(yù)覽,若沒有圖紙預(yù)覽就沒有圖紙。
  • 4. 未經(jīng)權(quán)益所有人同意不得將文件中的內(nèi)容挪作商業(yè)或盈利用途。
  • 5. 人人文庫網(wǎng)僅提供信息存儲(chǔ)空間,僅對(duì)用戶上傳內(nèi)容的表現(xiàn)方式做保護(hù)處理,對(duì)用戶上傳分享的文檔內(nèi)容本身不做任何修改或編輯,并不能對(duì)任何下載內(nèi)容負(fù)責(zé)。
  • 6. 下載文件中如有侵權(quán)或不適當(dāng)內(nèi)容,請(qǐng)與我們聯(lián)系,我們立即糾正。
  • 7. 本站不保證下載資源的準(zhǔn)確性、安全性和完整性, 同時(shí)也不承擔(dān)用戶因使用這些下載資源對(duì)自己和他人造成任何形式的傷害或損失。

評(píng)論

0/150

提交評(píng)論