ai/es (artificial intelligence / expert system) overview of ai

Download AI/ES (Artificial Intelligence / Expert System) Overview of AI

Post on 13-Jan-2016




5 download

Embed Size (px)


AI/ES (Artificial Intelligence / Expert System) Overview of AI. 2012. Fall. SME., Pukyong Nat ’ l Univ. Kim, Minsoo. Contents. What is AI? History of AI Research Area AI Systems. What is AI?. In the movies and novels, Faithful Servants & Friends Intelligent Machine  Autonomous Robot - PowerPoint PPT Presentation


  • AI/ES(Artificial Intelligence / Expert System)

    Overview of AI2012. Fall.SME., Pukyong Natl Univ.

    Kim, Minsoo

  • ContentsWhat is AI?History of AIResearch AreaAI Systems

  • What is AI?In the movies and novels,Faithful Servants & FriendsIntelligent Machine Autonomous RobotMetaphor for HumanismMan v.s. MachineIdentity ProblemDestruction v.s. New GenerationA Space Odyssey 2001Blade RunnerTerminatorI. RobotA.I.

  • What is AI?Human?Homo sapiens

    Man of wise Human intelligence

  • What is AI?AI Research AgendaProblem Solving with IntelligenceMotor Function Walking, Driving, Sensation & Perception OCR/OMR, Human-Like Problem SolvingDecision MakingHumanism or Humanoid(??)


  • What is AI?Four Different Definitions AIBehavior & Thinking ProcessExternal vs. Internal CharacteristicsReasoning (Ideal Logic vs. Rational Logic)Are Humans rational or irrational?1. Systems that think like humans(Cognitive Science)2. Systems that think rationally(Production Logics)3. Systems that act like humans(Turing Machine)4. Systems that act rationally(Intelligent Agents)IdealRationalThinkingBehavior

  • What is AI?Systems that think like humansCognitive Science ApproachMimic human thinking processBuild/Simulate computer model{ all inputs } AI system { Human-like outputs }1985, John HaugelandThe exciting new effort to make computers think machines with minds, in the full and literal sense.Artificial Intelligence: The Very Idea, MIT Press1978, Richard E. BellmanThe automation of activities that we associate with human thinking, activities such as decision-making, problem solving, learning An introduction to artificial intelligence: Can computers think?, Boyd & Fraser Publishing

  • What is AI?Systems that think rationallyInference Rule ApproachRational Thinking with Logical InferencingGreek Philosopher, Aristotle (Syllogistic logic)Socrates is a man; All men are mortal, therefore Socrates is mortal1985, E. Charniak & D. McDermottThe study of mental faculties through the use of computational modelsIntroduction to Artificial Intelligence, Addison-Wesley1992, P.H. WinstonThe study of the computations that make it possible to perceive, reason, and actArtificial Intelligence, Addison-Wesley

  • What is AI?Systems that act like humansTuring Test based ApproachCan machine think? Can machines do what we (as thinking entities) can do?Natural Language Processing, Knowledge Representation and Store, Automatic Inferencing, Pattern Recognition, Machine Learning, 1990, KurzweilThe art of creating machines that perform functions that require intelligence when performed by people1991, Rich & KnightThe study of how to make computers do things at which, at the moment, people are better

  • What is AI?Systems that act rationallyRational (Intelligent) Agent ApproachRational behavior: individuals maximize some objective function under the constraints (or under the uncertainty) they face. Exact inference is required but it cannot be always rational.There are cases when a simple reflex behavior is rational.More general and scientific approach1990, SchalkoffA field of study that seeks to explain and emulate intelligent behavior in terms of computational processes1993, Lugar & StubblefieldThe branch of computer science that is concerned with automation of intelligent behavior

  • History of AIThe Origins of AIAlan Turing30: A computer could exhibit intelligencebrilliant mathematicianWorked to crack German codes during WW2Worked on the development of the 1st computer that could store a program at Manchester UniversityThe Turing Test (1950)ability to achieve human-level performance, sufficient to fool an interrogator

  • History of AI1st Period, the dawn (1943~1951)1943, McCulloch & PittsDesign of Neural NetworkBrain Neuron Study, Propositional Logic, Turing TestLearning is required in the neurons network1949, Hebbs learning ruleEarly 1950s, Channon & TuringVon Neumann computer chess program1951, Minsky & EdmondDesigned SNARC(Stochastic Neural Analog Reinforcement Calculator)Randomly connected network of Hebb synapses (about 3000 of vacuum tubes and 40 neurons)

  • History of AI2nd Period, Early Study (1952~1965)Nowell & Simon, General Problem SolverModel human problem solving process Solve restricted puzzle (Tower of Hanoi)1958, MaCarthy (Dartmouth MIT)Develop LISPIntroduced Time Sharing SystemPaper: Programs with CommonsenseAdvice Taker: The first proposal to use logic to represent information in a computer.1958, MinskyMicroworlds problem solving (blocks world)Wide use of Neural Networks1962, Widrows Adaline (enhanced Hebbs learning rule)Rosenblatt, Perceptrons learning algorithm

  • History of AI3rd Period, Dark Era (1966~1974)1966, Negative report on machine translationDevastated natural language research for years1968, Marvin Minsky & Seymour PapertPinpoint the limitation of Perceptron NN researchs stagnationCause of depressionEarly AI programs somewhat lack domain knowledge and deliver information with just simple synaptic linksTackled somewhat complex problems from the beginningLimitations in their basic structure/frame for intelligent behavior

  • History of AI4th Period, Renaissance (1975~1990)General search problem domain specific search problem (with specialized knowledge)1975, Success on the Meta-Dendral project1980, Spotlight on the Expert SystemsMid-1980s, Return of NN w/ BackpropagationProsperous Era (1991 ~ )Wide variety of NN applications1990, Agent theoryAfter 2003, Information search Mobile Multi-Agent System

  • Research Area

  • Research AreaBasic Technology in AILearningInferenceRecognitionKnowledge BaseDatabaseLearning ModelInference EngineExpert SystemTheorem ProvingGameProblem SolvingChar/Doc/Voice/Image RecognitionNatural Lang.ProcessingPattern RecognitionSystemIntelligentSystem

  • AI SystemsWhat is AI Systems?Implement human mental modelSystem identification + System automation4 components of AI SystemUserHCI systemInference EngineKnowledge Base (RB + DB)ConsiderationsKn Definition: acquisition & understandingKn Representation: Semantics & ClassificationKn Manipulation: Reasoning, Control Strategy, Ambiguity Handling, Learning, Inferencing?Model Verification: Optimal? Available?

  • AI SystemsIn the end, AI system acquire knowledge, represent it internally, show the processed result to user via some interfaceProper application areasNo procedural algorithm exists, only heuristics existWhere human sensation and intuition works goodLimited knowledge workers, non-popular domainMedical, Law, Including uncertain information or dataReasonable level of data loss or existence of ambiguityDiagnosis, Inference, Prediction SystemFormal knowledge with few flexibility

  • AI SystemsConsiderations for applying AI systemDomain adequacyIn this domain proper to apply AI technique?Blind introduction can be more inefficientDoes it model the real system well?Is it truly a AI system?Is it efficient?


View more >