ARTIFICIAL INTELLIGENCE - FEUP
ARTIFICIAL INTELLIGENCEThinkARTIFICIAL INTELLIGENCEEugénio OliveiraAna Paula RochaHenrique Lopes CardosoSitio web institucionalSítio web específico: http://paginas.fe.up.pt/ eol/IA/ia1415.htmlEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEGENERIC GOALS OF THE COURSE:Going to a “new continent” of knowledge!3 key words:* KNOWLEDGE* LOGIC (more than Data)* ENGINEERING/ TECHNOLOGY (besides Science)Objectives:* TO LEARN* USE* DEVELOPConclusion:newotherdifferentMETHODS for problem solvingTECHNIQUES for implementing systemsPROGRAMS for solution searchTO LEARN new ways of solving Problems (usually complex)that need specific KNOWLEDGEEugénio Oliveira / FEUP
Artificial IntelligenceARTIFICIAL INTELLIGENCEI-INTRODUCTION TO ARTIFICIAL INTELLIGENCEMethods and Objectives:They are specific, although intersecting other areas: Computer Science, CognitiveScience, Neurosciences, Economics, Sociology, Psychology, Electronics, Scientific and technologic aspectsIn the domains of Programming, Algorithmic,Systems Analysis, Sensors andIother Engineering topics.Methodologic weaknessesLack of formalization for certain topics, regardig both theory and methodsproposed to design and implement “Intelligent” Systems.Eugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEARTIFICIAL INTELLIGENCE:Science Versus TechnologyCOMPUTER SCIENCECognitivePsycologySocionicsDecision TheoryGame TheoryARTIFICIAL icsResearch priorities for robustRoboticsand beneficialfor the last 20 yeartificial intelligars AI has beenence (S.Ru.sselfocused on theAgents - systemset al. Jan2015)problems surrothat perceive anunding the cond act in some enrelated to statisstructvironment. In thtical and econois context, the cr ion of intelligentmic notions ofplans, or infereiterion for intellrationality - collnces.igence isoquially, the ability to make good decisions,StatisticsEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEDefinitions" Artificial Intelligence studies those ideas that, when implementedin the computer make it possible to reach the same objectives thatmake people seem intelligent".“ More specifically, AI tries to make computers more useful and,simultaneously, studies those principles that make intelligence possible"Patrick Winston: ex-director of AI Lab at M.I.T.Comments ambiguous (definition contains what is being defined); "Lapalice-like" kind of statement (computers become more useful).Eugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCE“AI area presupposes the existence of processes on whichperception and reasoning rely and that these processes maybe scientifically studied and understood. Besides, it isabsolutely irrelevant for the theory of AI who (or what)“perceives” or “thinks” – man or computer. This is animplementation detail ”N. Nilsson ex-director of Stanford Research Institut; StanfordRobotics Lab.Comment: polemic. AI deviate , at least for a certain time period, fromthis paradigm in order to become more realistic and more independentfrom the way human brains work.NOW: Back to the fundamentalsEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCE"AI is the study of those processes that make it possible thecomputers to execute tasks for which, at the moment, people ismore effective.“E. Rich.Coment: Vague. Incomplete. But closer to the real truth in its simplicityEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEAnother Definition:Artificial Intelligence is a scientific discipline whose fundamental aimis to develop computational systems capable of showing operationalBehaviours that are similar to those of the humans in stereotypedsituations.Techniques of programming used, like non-deterministic search, arebased, at least partially, on declarative languages, namely logic-based,functional or at least, object-oriented.Eugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCE EVOLUTION:Cognitive Science Computer ScienceArtificial IntelligenceKnowledge EngineeringAutonomous AgentsApplicationDomainsRobotics DIFERENTIATION:Artificial Intelligence for:i) Complex problemsii) Reasoning ModelsEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEArtificial Intelligence Definitions organized in 4 categoriesThinkActMake Machinesthink Machines that executetasks that requireintelligencehumansComputational Modelsto study the rational mindTo study computationalprocesses to simulateintelligencerationalitySyst. that “Think” like HumansSyst. that “Think” rationallySyst. that “Act” like HumansSyst. that “Act” rationallyEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCECRONOLOGICAL 74PragmaticLogic Theorist and ics : Boole(IBM) Geometry P.S.FregePsycology: BehaviorismGame of ChekersCognitivism(A. Samuel)(experimentalisme W. James)LISP, Time-sharing(McCarthy)CyberneticsTuring testInformation Theory(Shannon)Search Knowledge(advice taker)Artificial Neuron ModelWhat is morepopular now in AI?Statistic-al Learning.You are killingyourself scientificallyResolution Principle(A.Robinson)Reaserch Groups:MIT (Minsky)Knowledge Engineering(Feigenbaum)DendralExpert Systems:MYCINUncertainty ReasoningProbabilistic :ProspectorFrames (Minsky)Pattern Recognition gets outHEARSAYII- Blackboard(McCulloch & Pits)Neural ComputationMinsky PhD thesis)Robotics("Shakey")Integral Calculus(SAINT)Grammar SIR (B.Raphael)Perceptron (Rosenblatt)ElizaMinsky & Papert andPerceptron polemicsPROLOG (Colmerauer)Tom Mitchell atStanford, “CdonceptsFormation (ML)Refounding gentsSoftware Agents"5th Generation ComputersºKISMET(MITI)RoboticHdw dedicated:Distributed AI Agents"Fuzzy“ controlMulti-Agent Syst.ESPRITCognitive AgNeural NetworksAI WebReactive Robotics(Brooks)KBSIn practiceSOAR - NewellMachine LearningCeasoning toRnamU.Stanford (McCarthy)uhrpedge, and SulewonKtrexpon Sense, EmmoCginBringAI is born:Meeting at Dartmouth College(Minsky, McCarthy, Simon,Newell, Shannon,.).2000Deep BlueNetworked/ SocialIntelligencesData & TextCOG at MITminingHumanoide Robot(R.Brooks)SemanticsFor NL ntsDeepLearning
ARTIFICIAL INTELLIGENCEINTRODUCTION TO ARTIFICIAL INTELLIGENCEI INTRODUCTIONII BASIC CONCEPTSSYLLABUSObjectiveMethodology (teaching and assessment)AI Evolution and ChronologyDocumentationDefinitions: what is AI?Applications: what domains?Agent Basic DefinitionsAgent Architectures : from Reactive to CognitiveIII PROBLEM SOLVING METHODS"Production“ SystemsControl Strategies for Systematic SearchForward and Backwards Chaining (Depth-first and Breadth-first)Irrevocable Search: "hill climbing“ and “Simulated Annealing”Search by trial: "backtracking; Graph search“Branch and Bound”Eugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEIII PROBLEM SOLVING METHODS (cont.)Evolutionary Computation (Genetic Algorithms)Heuristic Search : “Best-First"A* Algorithm and admissibilityMeans-Ends AnalysisConstraint Satisfaction: "Relaxation“ Principles** Search for “Games”: Minmax algorithmAlfa-Beta pruningProlog examples of basic methods:Interpreters breadth-first and depth-first** Students Oral presentation and mini-test assessment in the classEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEIV INTRODUCTION TO KNOWLEDGE REPRESENTATIONRepresentation Systems definitionStructures for representing knowledge:Production Rules; Assotiative Networks (Semantic Networks)"Frames";Predicate Logic; Other LogicsUncertainty Reasoning:Probabilistic Models;Certainty Factors;Dempster-Schafer Model;** Fuzzy Sets LogicLogics: Propositional Logic, Predicate Logic; Intentional Logic (mention)** Students Oral presentation and mini-test assessment in the classV KNOWLEDGE ENGINEERINGKBSExpert Systems:CharaterizationArchitectureKnowledge Representation and Meta-knowledgeInference Motor and Explanation GenerationExpert Systems Case studies:ORBI; SMYCIN; ARCADemonstrationsGeneric Systems: "Shells"Eugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEVI INTRODUCTION TO COMPUTATIONAL NATURAL LANGUAGEObjectives and difficultiesSyntactic and semantic AnalysisATN; Semantic Networks; "Frames“; typical cases (mention)Classic approach and use of Logic:Definite Clause Grammars; a few examples in PortugueseExtraposition GrammarsVIIINTRODUCTION TO MACHINE LEARNINGLearning modesConcepts learning; by example; by analogyExplanation Based Learning (EBL) :Algorithms for EBG, mEBG and IOL;ExamplesInductive Learning: Algorithms ID3 and C4.5Application ExamplesVIII INTRODUCTION TO NEURAL NETWORKSBasic Principles and Concepts** fundamental AlgorithmsApplication Example** Students Oral presentation and mini-test assessment in the classEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCESCHEDULEIntr. AIBásicconcepts12Problem Solving MethodsBlind search.3Homeassignment45GA6GameHeuristic SearhT.sear.78910 11Expert NaturalSystem mb. MachineLearning20Knowledge Rep; ESAgents, search, minimax, GA, optimizationExercices in Prolog and Projects2122NaturalLang.ANN2324MLearningEugénio Oliveira / FEUP25 26 27
ARTIFICIAL INTELLIGENCEBIBLIOGRAPHY for Artificial IntelligenceNotes available at the course web site- Eugénio OliveiraBOOKS: "ARTIFICIAL INTELLIGENCE: A Modern Approach", S.Russel and P.Norvig; Prentice Hall, 3rd Ed 2010 "ARTIFICIAL INTELLIGENCE" E. Rich; K. Knight, 2nd Ed., MacGraw-Hill, 1991 “C4.5-Programs for Machine Learning" Ross Quinlan,Morgan Kaufmann,1993 "THE ART OF PROLOG" Sterling and Shapiro, MIT Press, 1986 "INTELIGÊNCIA ARTIFICIAL-Fundamentos e Aplicações " E.Costa, A.Simões; FCA editores, 2004 JOURNALS: “Autonomous Agents and Multi-Agent Sytems”, Springer "ARTIFICIAL INTELLIGENCE“ Elsevier-North-Holland "IEEE EXPERT" "MACHINE LEARNING" Kluwer A.P.Eugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEMini-Projects home assignmentsA. Pesquisa Sistemática/Informada de Soluções (SEARCH)A1. Pesquisa de trajetos em redes de transportes públicosA2. Trajeto de um robô em ambiente conhecidoA3. Pesquisa aplicada ao Problema de Alocação de Lotes de TerrenoA4. Pesquisa aplicada à resolução do jogo RushA5. Pesquisa aplicada à resolução do Solitário SokobanMétodos: Pesquisa em Profundidade, Largura, Profundidade Iterativa, Bidireccional,Gulosa, Algoritmo A*, Heurísticas B. Resolução de Problemas de Otimização(Optimization)B1. Otimização de Corte de Placas de madeira/vidroB2. Otimização do Problema de Alocação de Lotes de TerrenoB3. Otimização de Horários de Motoristas dos STCPB4. Otimização da Localização de Prontos-Socorro numa CidadeB5. Aplicação de Algoritmos Genéticos para localização de uma BarragemMétodos: Pesquisa Sistemática/Informada, Algoritmos Genéticos, PesquisaTabu, Arrefecimento SimuladoEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEMini-Projects home assignmentsC. Games Tree SearchC1. Jogo de Tabuleiro – Tic-Tac-KuC2. Jogo de Tabuleiro – YinshC3. Jogo de Tabuleiro – HexC4. Jogo de Tabuleiro – BlockadeC5. Jogo de Tabuleiro – Samurai de Reiner KniziaMétodos: Algoritmo MiniMax com Cortes Alfa-Beta e variações desteD. Knowledge Engineering and Natural LanguageD1. Desenvolvimento de uma "Shell" (com fuzzy)D2. Sistema de Regras para controlo de dispositivos de Domótica, usando JessD3. Informações sobre voos da TAP em Linguagem NaturalD4. Informações sobre Filmes de Cinema em Cartaz em Linguagem NaturalD5. Informações sobre Restaurantes na cidade do Porto em Linguagem NaturalMétodos: Representação do Conhecimento, Raciocínio Incerto, Sistemas Periciais,Linguagem NaturalEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEMini-Projects home assignmentsE. Machine Learning and Artificial Neural NetworksE1. Reconhecimento de Sinais de Trânsito utilizando Redes NeuronaisE2. Aplicação de ID3 ou C4.5 à classificação de Área Destruída em IncêndioE3. Previsão de Área destruída num incêndio utilizando Redes NeuronaisE4. Aplicação de ID3 ou C4.5 à classificação da Qualidade de Vinhos VerdesE5. Previsão da qualidade de um Vinho Verde utilizando Redes NeuronaisMétodos: Algoritmos de Aprendizagem ID3 e C4.5, Redes Neuronais ArtificiaisEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEGRADING Grading calculation:– Exam50% (minimum 7,5 in 20)– Continuous learning»»»»50% % (minimum 7,5 in 20)Report Intermediate workFinal ReportParticipation and mini-tests in the classesFinal home work presentation15%15%30%40%Eugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEGRADING Course effort in ECTS 6 ECTS 1 ECTS 26-27H Total 156-162 H Classes (T TP): Project: Study Exam preparation64h (4h*16s)50h44hEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCETECHNICSSIMPLE EXAMPLES and intuitive approaches “AIoriented” (GOFAI):A- search:“Tic-Tac-Toe” GameTraditional approach:-Simple algorithm-DisadvantagesnAdva100201000O XOMatrix M with all possible configurations as vetors (Vi)Ternary number (configuration) à decimalIndexing a new position in the MatrixNew position corresponds to the vector result VjNeeds memory; introduce all the situationsà errors;unflexible- New approach:- Algorithm:stCalculatepossible configurations (sons of a node)panextehthtiarn w the solution OTHERWISE consider possible answers for each nodefinded: lec- Disadvantage: More time (all possible sequences before each movement)-Advantage: Extensible, versatile (different games, heuristics; several levels), less memoryEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCETECHNICSExample B: Knowledge RepresentationLetters Recognition- A traditional approach:analogic matrix : Inputs: counting 0s and 1s from the analogic matrix“hashing” table to index patterns(key: nº 1s in 3 lines of the matrix and combine them without collisions in a hashing f.)B- Disadvantages:I J Are Deviations in letters meaningful or not?Too many patterns imply collisionslI- An alternative:- Count the nº of 1s in each sub-area (or ratio between “1” and “0”)- Build a vector with the results- Compute the distance to each patternEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCETECHNICS- New approach:Ne-Patterns as sets of characteristics description: he use oftcludnisthodemrOtheeacbsworkteNu r alarc (a,b) AND up(a,b) AND line(b,c) AND left(c,b) AND(nil OR (line(b,d) AND down(d,b))) AND(nil OR (line(a,e) AND up(e,a)))d- Algorithm - Find instances of Primitives(arcs, lines)- Relate them and compare with existent patterns- Select the closer pattern-Advantage Permits size variation; alternative descriptions become simple-AI TECHNICS involved in the examples: SEARCH and SIYMBOLIC REPRESENTATIONEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEAI Application and Research DomainsAi RESEARCH AND APPLICATION DOMAINS1 Computational Natural Languageunderstand translate between structures superficial understanding : Lexical Analysis (Eg: Eliza ) Deeper Analysis implies:cesSintax, Semantics, PragmaticsoccurrenocdnaquencySentence: John kicks the white ballords frewnodesS-- NS VSNS -- Qt NS1NS -- NS1NS1 -- N ADJSADJS-- nil ADJ ADJSVS -- V NSN-- John ball ADJ -- white Qt -- the V-- kickse baaches arorppawNeelemental GrammarEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEAI Application and Research Domains Semantic Interpretation is fundamental to select the Actionsex: What is the value of soil Aptitude to Agriculture at coordinatesX1,Y1?After Syntactic Analysis, the semantic analysis generates:coordinates(X1, Y1, D, V), a( 3, D, V-R). The Action is the answer to the defined questions (sub-goals)Other situations requiring semantic analysis:The politician lied about the subject X à negative opinion about the polititianethodsmtneifficmore eeraerTheEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEAI Application and Research Domains"EXPERT SYSTEMS"SPECIFIC VS GENERALKnowledge Vs "Intelligence"SYSTEM THAT SIMULATES THE EXPERT:Using SYMBOLIC AND HEURISTIC KnowledgeUsing de UNCERTAIN, VAGUE, INCOMPLETE KnowledgeMODULAR access to KnowledgeEXPLANATIONS capabilitiesKnowledge ACQUISITION (might be automatic)Eugénio Oliveira / FEUP
Domínios paraa INTELIGÊNCIAARTIFICIALINTELLIGENCE ARTIFICIALROBOTICS- ARCHITECTURES: Cognitive and Reactive; Hybrid- PERCEPTION: Scenes Interpretation- DECISION: Planning. "Frame Problem"- Languages: Task Level- Vision Modelling Interpretation- Teams COORDINATIONEugénio Oliveira / FEUP
Domínios paraa INTELIGÊNCIAARTIFICIALINTELLIGENCE ARTIFICIALLEARNING, ADAPTATION and “DATA MINING”* Induction of Rules based in:- analogy; examples; explanations* “Data e Text Mining”- Generation of new Knowledge;- Patterns Recognition (text, image, music )- Opinion Extraction, Summarization* progressive Adaptation (Evolutionary Algorithms)NEW LOGICS for Automatic Reasoning- Order N- Modal and Intensional- temporal- non-monotonicEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCECONETIVISTE and NEURAL Architectures- sub-simbolic Information : Preview Adaptive Control Paattern RecognitionDISTRIBUTED ARTIFICIAL INTELLIGENCE-DISTRIBUTED AND COOPERATIVE AGENTSAplications in domains DDD:- Networks management and analysis- Shop floors (CIM)- Softbots (Shopbots,.)- Electronic Markets (Auctions, contracts)- Electronic Institutions (Virtual EnterprisesNegotiation, contracting)- “Emotionl” Agents- Simulation for traffic,.Eugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEDomains for the ARTIFICIAL INTELLIGENCEINTELLIGENT TUTORSKnowledge Representation of the domain,Pedagogical strategiesAdaptation/Classification and profile extractionSIMULATION of Human BEHAVIOURSArchitectures type: “mentalistic” and based in “Emotions”Team Coordination of autonomous entitiesecological SystemsECONOMIC COMPUTATION BASED IN AGENTS (ACE)“Computational study of economic processes as dynamicsystems of interacting agents”Eugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEComputational AGENTS :a) DefinitionsEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEAgent DefinitionsElementary Notions : Agents are computational entities with the capability of perceivingthe external environment (through Sensors) as well as interacting withthis environment through “effectors”). Agents make possible to humans to “delegate” to them responsibilitiesthat are both costly in time and “power” (computation, memory,Communication, repeatability.) Agents use perception sequences together with a priori knowledge inorder to select actions maximizing their performanceAgent Platform (Communication and Distribution) ProgramProgram Architecture Modules’ ProgramsEugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEAgents Definitions Agent Definition: Computing Entity controlled by a program which is situatedin an Environment and capable of deciding with autonomy inthatEnvironment while pursuing its Goals Autonomy Reactivity Pro-activity Sociability Agents Vs Objects: Agents are autonomous: They may refuse demands. Objects no! Objects control their state, not their behaviourAgents control their state and their behaviourEugénio Oliveira / FEUP
Agent DefinitionsARTIFICIAL INTELLIGENCE Strong Definition: Autonomous computational Systems, pro-active, whose reasoningprocess is based on “mentalistic” concepts such as: Beliefs, Knowledge, Intentions, Commitment, Goals,Emotions. Other possible Attributes : Mobility Benevolence Rationality / “Emotionality”Eugénio Oliveira / FEUP
ARTIFICIAL INTELLIGENCEAgent DefinitionsAgent "PAGE" Description:[Perception, Actions, "Goals" (objectives), Environment]Now (PEAS- Performance measures, Environment, Actuators, Sensors)Agent ExamplesTypePerception ActionsGoalsMedical Diagnostic Symptoms graphicSoftbot orShopBotsEnvironmentPatientPacient signalshealthCostsHospitalMinimization instrumentationpixelsintensityPick upPut downpartsLocate andplacepartsTablesConveyorbelts .Finding andReadingWeb ionSelectionComputersInternetWeb pagesEugénio Oliveira / FEUP
Agent DefinitionsARTIFICIAL INTELLIGENCEIs it dif
ARTIFICIAL INTELLIGENCE Eugénio Oliveira / FEUP ARTIFICIAL INTELLIGENCE Eugénio Oliveira . Techniques of programming used, like non-deterministic search, are based, . Networked / Social Intelligence s Data & Text mining Semantics For NL and, Web E- Business Intelligenc e Mentalisti c Agents
Artificial Intelligence -a brief introduction Project Management and Artificial Intelligence -Beyond human imagination! November 2018 7 Artificial Intelligence Applications Artificial Intelligence is the ability of a system to perform tasks through intelligent deduction, when provided with an abstract set of information.
and artificial intelligence expert, joined Ernst & Young as the person in charge of its global innovative artificial intelligence team. In recent years, many countries have been competing to carry out research and application of artificial intelli-gence, and the call for he use of artificial
BCS Foundation Certificate in Artificial Intelligence V1.1 Oct 2020 Syllabus Learning Objectives 1. Ethical and Sustainable Human and Artificial Intelligence (20%) Candidates will be able to: 1.1. Recall the general definition of Human and Artificial Intelligence (AI). 1.1.1. Describe the concept of intelligent agents. 1.1.2. Describe a modern .
IN ARTIFICIAL INTELLIGENCE Stuart Russell and Peter Norvig, Editors FORSYTH & PONCE Computer Vision: A Modern Approach GRAHAM ANSI Common Lisp JURAFSKY & MARTIN Speech and Language Processing, 2nd ed. NEAPOLITAN Learning Bayesian Networks RUSSELL & NORVIG Artificial Intelligence: A Modern Approach, 3rd ed. Artificial Intelligence A Modern Approach Third Edition Stuart J. Russell and Peter .
Peter Norvig Prentice Hall, 2003 This is the book that ties in most closely with the module Artificial Intelligence (2nd ed.) Elaine Rich & Kevin Knight McGraw Hill, 1991 Quite old now, but still a good second book Artificial Intelligence: A New Synthesis Nils Nilsson Morgan Kaufmann, 1998 A good modern book Artificial Intelligence (3rd ed.) Patrick Winston Addison Wesley, 1992 A classic, but .
BCS Essentials Certificate in Artificial Intelligence Syllabus V1.0 BCS 2018 Page 10 of 16 Recommended Reading List Artificial Intelligence and Consciousness Title Artificial Intelligence, A Modern Approach, 3rd Edition Author Stuart Russell and Peter Norvig, Publication Date 2016, ISBN 10 1292153962
PA R T 1 Introduction to Artificial Intelligence 1 Chapter 1 A Brief History of Artificial Intelligence 3 1.1 Introduction 3 1.2 What Is Artificial Intelligence? 4 1.3 Strong Methods and Weak Methods 5 1.4 From Aristotle to Babbage 6 1.5 Alan Turing and the 1950s 7 1.6 The 1960s to the 1990s 9 1.7 Philosophy 10 1.8 Linguistics 11
CISC4/681 Introduction to Artificial Intelligence 1 Russell and Norvig: 2 Agents? agent percepts sensors actions environment CISC4/681 Introduction to Artificial Intelligence 2 Agent – perceives the environment through sensors and acts on it through actuators Percept – agent’s perceptual input (the basis for its actions) Percept Sequence – complete history of what has been .
