Slide 1

Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley Chapter 11: Artificial Intelligence Computer Science: An Overview Tenth Edition by J. Glenn Brookshear Modified by Senem Kumova Metin Slide 2 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-2 Chapter 11: Artificial Intelligence 11.1 Intelligence and Machines 11.2 Perception (Understanding Image, Speech or WrittenText ) 11.3 Reasoning Slide 3 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-3 Intelligent Agents Agent: A device that responds to stimuli from its environment Sensors Actuators (e.g. wheels,legs, speech sythesizers) Much of the research in artificial intelligence can be viewed in the context of building agents that behave intelligently Slide 4 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-4 Levels of Intelligent Behavior Reflex: predetermined responses to the input data More intelligent behavior requires knowledge of the environment and involves such activities as: Goal seeking (e.g Winning a game of chess) Learning (developing procedural knowledge or storing declarative knowledge) Slide 5 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-5 Figure 11.1 The eight-puzzle in its solved configuration Slide 6 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-6 Figure 11.2 Our puzzle-solving machine (The agent) Slide 7 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-7 Approaches to Research in Artificial Intelligence Engineering track (develop systems that exhibit intelligent behaviour) Performance oriented Theoretical track (develop a computational understanding of human behaviour) Simulation oriented Slide 8 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-8 Turing Test Test setup: Human interrogator communicates with test subject by typewriter. Test: Can the human interrogator distinguish whether the test subject is human or machine? Slide 9 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-9 Techniques for Understanding Images Template matching Image processing (identifying characteristics of images) edge enhancement region finding smoothing, Image analysis (understanding what these characterictics mean) Slide 10 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley Image processing Edge enhancement : Applying techniques to fclarify boundaries between regions Region finding : Find areas which have common properties Smoothing: Removing flaws in the image 11-10 Slide 11 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley Smoothing 11-11 Slide 12 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-12 Language Processing Syntactic Analysis (e.g. Parsing, Tokenization) Mary gave a card Subject verb object Semantic Analysis -> Find grammatical role of each word Mary gave John a card = John got a card from Mary Contextual Analysis (understanding process) Kitaplar tarken kolum koptu Slide 13 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley Language Processing Information retrieval : Identify the documents related to the topic at hand Information extraction: Extracting information from documents so that it takes a form that is useful in other applications --Answers of questions are stored as forms 11-13 Slide 14 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-14 Figure 11.3 A semantic net (A type of form which information extractor record information) Slide 15 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-15 REASONING: Components of a Production Systems 1. Collection of states Start (or initial) state Goal state (or states) 2. Collection of productions(rules or moves) Production: an operation that can be performed in the applcation environment to move from one state to another Each production may have preconditions 3. Control system: decides which production to apply next (it includes the problem space) Slide 16 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-16 Reasoning by Searching State Graph: All states and productions Search Tree: A record of state transitions explored while searching for a goal state Breadth-first search Depth-first search Slide 17 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-17 Figure 11.4 A small portion of the eight-puzzles state graph Slide 18 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-18 Figure 11.5 Deductive reasoning in the context of a production system (Inference rules) Slide 19 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-19 Figure 11.6 An unsolved eight-puzzle Slide 20 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-20 Figure 11.7 A sample search tree Slide 21 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-21 Figure 11.8 Productions stacked for later execution Slide 22 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-22 Heuristic Strategies Developing a full search tree is impratical in some cases, as a result instead of building the tree in a breath first manner, build it using depth first method. Identify which of several steps to be the most promising Heuristic: A rule of thumb for making decisions Slide 23 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-23 Heuristic Strategies Heuristic: A rule of thumb for making decisions Requirements for good heuristics Must be easier to compute than a complete solution Must provide a reasonable estimate of proximity to a goal Slide 24 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-24 Figure 11.9 An unsolved eight-puzzle Slide 25 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-25 Figure 11.10 An algorithm for a control system using heuristics Slide 26 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-26 Figure 11.11 The beginnings of our heuristic search Slide 27 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-27 Figure 11.12 The search tree after two passes Slide 28 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-28 Figure 11.13 The search tree after three passes Slide 29 Copyright 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley 11-29 Figure 11.14 The complete search tree formed by our heuristic system