Dr. Peter Norvig, contributing Artificial Intelligence author and Professor Sebastian Thrun, a Pearson author are offering a free online course at Stanford University on artificial intelligence. According to an article in The New York Times , the course on artificial intelligence is “one of three being offered experimentally by the Stanford computer science department to extend technology knowledge and skills beyond this elite campus to the entire world.” One of the other two courses, an introduction to database software, is being taught by Pearson author Dr. Jennifer Widom.
Artificial Intelligence A Modern Approach Stuart J Russell and Peter Norvig Contributing writers: John F Canny, Jitendra M Malik, Douglas D Edwards Prentice Hall, Englewood Cliffs, New Jersey 07632 Library of Congress Cataloging-in-Publication Data Russell, Stuart J (Stuart Jonathan) Artificial intelligence : a modern approach/ Stuart Russell, Peter Norvig p cm Includes bibliographical references and index ISBN 0-13-103805-2 Artificial intelligence I Norvig, Peter II Title Q335.R86 1995 006.3-dc20 94-36444 CIP Publisher: Alan Apt Production Editor: Mona Pompili Developmental Editor: Sondra Chavez Cover Designers: Stuart Russell and Peter Norvig Production Coordinator: Lori Bulwin Editorial Assistant: Shirley McGuire © 1995 by Prentice-Hall, Inc A Simon & Schuster Company Englewood Cliffs, New Jersey 07632 The author and publisher of this book have used their best efforts in preparing this book These efforts include the development, research, and testing of the theories and programs to determine their effectiveness The author and publisher shall not be liable in any event for incidental or consequential damages in connection with, or arising out of, the furnishing, performance, or use of these programs All rights reserved No part of this book may be reproduced, in any form or by any means, without permission in writing from the publisher Printed in the United States of America 10 ISBN D-IH-IQBSOS-E Prentice-Hall International (UK) Limited, London Prentice-Hall of Australia Pty Limited, Sydney Prentice-Hall Canada, Inc., Toronto Prentice-Hall Hispanoamericana, S.A., Mexico Prentice-Hall of India Private Limited, New Delhi Prentice-Hall of Japan, Inc., Tokyo Simon & Schuster Asia Pte Ltd., Singapore Editora Prentice-Hall Brasil, Ltda., Rio de Janeiro Preface There are many textbooks that offer an introduction to artificial intelligence (AI) This text has five principal features that together distinguish it from other texts Unified presentation of the field Some texts are organized from a historical perspective, describing each of the major problems and solutions that have been uncovered in 40 years of AI research Although there is value to this perspective, the result is to give the impression of a dozen or so barely related subfields, each with its own techniques and problems We have chosen to present AI as a unified field, working on a common problem in various guises This has entailed some reinterpretation of past research, showing how it fits within a common framework and how it relates to other work that was historically separate It has also led us to include material not normally covered in AI texts Intelligent agent design The unifying theme of the book is the concept of an intelligent agent In this view, the problem of AI is to describe and build agents that receive percepts from the environment and perform actions Each such agent is implemented by a function that maps percepts to actions, and we cover different ways to represent these functions, such as production systems, reactive agents, logical planners, neural networks, and decision-theoretic systems We explain the role of learning as extending the reach of the designer into unknown environments, and show how it constrains agent design, favoring explicit knowledge representation and reasoning We treat robotics and vision not as independently defined problems, but as occurring in the service of goal achievement We stress the importance of the task environment characteristics in determining the appropriate agent design Comprehensive and up-to-date coverage We cover areas that are sometimes underemphasized, including reasoning under uncertainty, learning, neural networks, natural language, vision, robotics, and philosophical foundations We cover many of the more recent ideas in the field, including simulated annealing, memory-bounded search, global ontologies, dynamic and adaptive probabilistic (Bayesian) networks, computational learning theory, and reinforcement learning We also provide extensive notes and references on the historical sources and current literature for the main ideas in each chapter Equal emphasis on theory and practice Theory and practice are given equal emphasis All material is grounded in first principles with rigorous theoretical analysis where appropriate, but the point of the theory is to get the concepts across and explain how they are used in actual, fielded systems The reader of this book will come away with an appreciation for the basic concepts and mathematical methods of AI, and also with an idea of what can and cannot be done with today's technology, at what cost, and using what techniques Understanding through implementation The principles of intelligent agent design are clarified by using them to actually build agents Chapter provides an overview of agent design, including a basic agent and environment vii Preface Vlll project Subsequent chapters include programming exercises that ask the student to add > capabilities to the agent, making it behave more and more interestingly and (we hope) intelligently Algorithms are presented at three levels of detail: prose descriptions and ! pseudo-code in the text, and complete Common Lisp programs available on the Internet or on floppy disk All the agent programs are interoperable and work in a uniform framework for simulated environments This book is primarily intended for use in an undergraduate course or course sequence It can also be used in a graduate-level course (perhaps with the addition of some of the primary sources suggested in the bibliographical notes) Because of its comprehensive coverage and the large number of detailed algorithms, it is useful as a primary reference volume for AI graduate students and professionals wishing to branch out beyond their own subfield We also hope that AI researchers could benefit from thinking about the unifying approach we advocate The only prerequisite is familiarity with basic concepts of computer science (algorithms, data structures, complexity) at a sophomore level Freshman calculus is useful for understanding neural networks and adaptive probabilistic networks in detail Some experience with nonnumeric programming is desirable, but can be picked up in a few weeks study We provide implementations of all algorithms in Common Lisp (see Appendix B), but other languages such as Scheme, Prolog, Smalltalk, C++, or ML could be used instead Overview of the book The book is divided into eight parts Part 1, "Artificial Intelligence," sets the stage for all the others, and offers a view of the AI enterprise based around the idea of intelligent agents—systems that can decide what to and it Part II, "Problem Solving," concentrates on methods for deciding what to when one needs to think ahead several steps, for example in navigating across country or playing chess Part III, "Knowledge and Reasoning," discusses ways to represent knowledge about the world—how it works, what it is currently like, what one's actions might do—and how to reason logically with that knowledge Part IV, "Acting Logically," then discusses how to use these reasoning methods to decide what to do, particularly by constructing plans Part V, "Uncertain Knowledge and Reasoning," is analogous to Parts III and IV, but it concentrates on reasoning and decision-making in the presence of uncertainty about the world, as might be faced, for example, by a system for medical diagnosis and treatment Together, Parts II to V describe that part of the intelligent agent responsible for reaching decisions Part VI, "Learning," describes methods for generating the knowledge required by these decision-making components; it also introduces a new kind of component, the neural network, and its associated learning procedures Part VII, "Communicating, Perceiving, and Acting," describes ways in which an intelligent agent can perceive its environment so as to know what is going on, whether by vision, touch, hearing, or understanding language; and ways in which it can turn its plans into real actions, either as robot motion or as natural language utterances Finally, Part VIII, "Conclusions," analyses the past and future of AI, and provides some light amusement by discussing what AI really is and why it has already succeeded to some degree, and airing the views of those philosophers who believe that AI can never succeed at all Preface Using this book This is a big book; covering all the chapters and the projects would take two semesters You will notice that the book is divided into 27 chapters, which makes it easy to select the appropriate material for any chosen course of study Each chapter can be covered in approximately one week Some reasonable choices for a variety of quarter and semester courses are as follows: • One-quarter general introductory course: Chapters 1, 2, 3, 6, 7, 9, 11, 14, 15, 18, 22 • One-semester general introductory course: Chapters 1, 2, 3, 4, 6, 7, 9, 11, 13, 14, 15, 18, 19, 22, 24, 26, 27 • One-quarter course with concentration on search and planning: Chapters 1, 2, 3, 4, 5, 6, 7, 9, 11, 12,13 • One-quarter course with concentration on reasoning and expert systems: Chapters 1,2, 3, 6, 7, 8,9, 10,11,14, 15,16 • One-quarter course with concentration on natural language: Chapters 1, 2, 3, 6, 7, 8, 9, 14, 15, 22, 23, 26, 27 • One-semester course with concentration on learning and neural networks: Chapters 1, 2, 3, 4, 6, 7, 9, 14, 15, 16, 17,18, 19, 20, 21 • One-semester course with concentration on vision and robotics: Chapters 1, 2, 3, 4, 6, 7, 11, 13, 14, 15, 16, 17, 24, 25, 20 These sequences could be used for both undergraduate and graduate courses The relevant parts of the book could also be used to provide the first phase of graduate specialty courses For example, Part VI could be used in conjunction with readings from the literature in a course on machine learning We have decided not to designate certain sections as "optional" or certain exercises as "difficult," as individual tastes and backgrounds vary widely Exercises requiring significant programming are marked with a keyboard icon, and those requiring some investigation of the literature are marked with a book icon Altogether, over 300 exercises are included Some of them are large enough to be considered term projects Many of the exercises can best be solved by taking advantage of the code repository, which is described in Appendix B Throughout the book, important points are marked with a pointing icon If you have any comments on the book, we'd like to hear from you Appendix B includes information on how to contact us Acknowledgements Jitendra Malik wrote most of Chapter 24 (Vision) and John Canny wrote most of Chapter 25 (Robotics) Doug Edwards researched the Historical Notes sections for all chapters and wrote much of them Tim Huang helped with formatting of the diagrams and algorithms Maryann Simmons prepared the 3-D model from which the cover illustration was produced, and Lisa Marie Sardegna did the postprocessing for the final image Alan Apt, Mona Pompili, and Sondra Chavez at Prentice Hall tried their best to keep us on schedule and made many helpful suggestions on design and content Preface Stuart would like to thank his parents, brother, and sister for their encouragement and their patience at his extended absence He hopes to be home for Christmas He would also like to thank Loy Sheflott for her patience and support He hopes to be home some time tomorrow afternoon His intellectual debt to his Ph.D advisor, Michael Genesereth, is evident throughout the book RUGS (Russell's Unusual Group of Students) have been unusually helpful Peter would like to thank his parents (Torsten and Gerda) for getting him started, his advisor (Bob Wilensky), supervisors (Bill Woods and Bob Sproull) and employer (Sun Microsystems) for supporting his work in AI, and his wife (Kris) and friends for encouraging and tolerating him through the long hours of writing Before publication, drafts of this book were used in 26 courses by about 1000 students Both of us deeply appreciate the many comments of these students and instructors (and other reviewers) We can't thank them all individually, but we would like to acknowledge the especially helpful comments of these people: Tony Barrett, Howard Beck, John Binder, Larry Bookman, Chris Brown, Lauren Burka, Murray Campbell, Anil Chakravarthy, Roberto Cipolla, Doug Edwards, Kutluhan Erol, Jeffrey Forbes, John Fosler, Bob Futrelle, Sabine Glesner, Barbara Grosz, Steve Hanks, Othar Hansson, Jim Hendler, Tim Huang, Seth Hutchinson, Dan Jurafsky, Leslie Pack Kaelbling, Keiji Kanazawa, Surekha Kasibhatla, Simon Kasif, Daphne Roller, Rich Korf, James Kurien, John Lazzaro, Jason Leatherman, Jon LeBlanc, Jim Martin, Andy Mayer, Steve Minton, Leora Morgenstern, Ron Musick, Stuart Nelson, Steve Omohundro, Ron Parr, Tony Passera, Michael Pazzani, Ira Pohl, Martha Pollack, Bruce Porter, Malcolm Pradhan, Lorraine Prior, Greg Provan, Philip Resnik, Richard Scherl, Daniel Sleator, Robert Sproull, Lynn Stein, Devika Subramanian, Rich Sutton, Jonathan Tash, Austin Tate, Mark Torrance, Randall Upham, Jim Waldo, Bonnie Webber, Michael Wellman, Dan Weld, Richard Yen, Shlomo Zilberstein Summary of Contents i Artificial Intelligence ii 1 Introduction Intelligent A g e n t s 31 Problem-solving 53 Solving Problems by Searching 55 Informed Search Methods 92 Game P l a y i n g 122 in Knowledge and reasoning 10 Acting logically IV 149 Agents that Reason L o g i c a l l y 151 First-Order L o g i c 185 Building a Knowledge Base 217 Inference in First-Order L o g i c 265 Logical Reasoning S y s t e m s 297 335 11 P l a n n i n g 337 12 Practical Planning 367 13 Planning and A c t i n g 392 Uncertain knowledge and reasoning 413 14 U n c e r t a i n t y 415 15 Probabilistic Reasoning S y s t e m s 436 16 Making Simple Decisions 471 17 Making Complex Decisions 498 VI Learning 18 19 20 21 VII Communicating, perceiving, and acting 22 23 24 25 VIII 649 Agents that Communicate 651 Practical Natural Language Processing 691 Perception 724 R o b o t i c s 773 Conclusions 26 27 A B 523 Learning from O b s e r v a t i o n s 525 Learning in Neural and Belief N e t w o r k s 563 Reinforcement L e a r n i n g 598 Knowledge in L e a r n i n g 625 815 Philosophical Foundations 817 AI: Present and Future 842 Complexity analysis and O() n o t a t i o n 851 Notes on Languages and A l g o r i t h m s 854 Bibliography Index 859 905 Contents I Artificial Intelligence 1 Introduction 1.1 What is AI? Acting humanly: The Turing Test approach Thinking humanly: The cognitive modelling approach Thinking rationally: The laws of thought approach Acting rationally: The rational agent approach 1.2 The Foundations of Artificial Intelligence Philosophy (428 B.C.-present) Mathematics (c 800-present) Psychology (1879-present) Computer engineering (1940-present) Linguistics (1957-present) 1.3 The History of Artificial Intelligence The gestation of artificial intelligence (1943-1956) Early enthusiasm, great expectations (1952-1969) A dose of reality (1966-1974) Knowledge-based systems: The key to power? (1969-1979) AI becomes an industry (1980-1988) The return of neural networks (1986-present) Recent events (1987-present) 1.4 The State of the Art 1.5 Summary Bibliographical and Historical Notes Exercises 6 8 11 12 14 15 16 16 17 20 22 24 24 25 26 27 28 28 Intelligent Agents 2.1 Introduction 2.2 How Agents Should Act The ideal mapping from percept sequences to actions Autonomy 2.3 Structure of Intelligent Agents Agent programs Why not just look up the answers? An example Simple reflex agents 31 31 31 34 35 35 37 38 39 40 Agents that keep track of the world Goal-based agents Utility-based agents Environments 41 42 44 45 2.4 Contents XIV Properties of environments Environment programs 2.5 Summary Bibliographical and Historical Notes Exercises 46 47 49 50 50 II Problem-solving 53 Solving Problems by Searching 3.1 Problem-Solving Agents 3.2 Formulating Problems Knowledge and problem types Well-defined problems and solutions Measuring problem-solving performance Choosing states and actions 3.3 Example Problems Toy problems Real-world problems 3.4 Searching for Solutions Generating action sequences Data structures for search trees 3.5 Search Strategies Breadth-first search Uniform cost search Depth-first search Depth-limited search Iterative deepening search Bidirectional search Comparing search strategies 3.6 Avoiding Repeated States 3.7 Constraint Satisfaction Search 3.8 Summary Bibliographical and Historical Notes Exercises 55 55 57 58 60 61 61 63 63 68 70 70 72 73 74 75 77 78 78 80 81 82 83 85 86 87 Informed Search Methods 92 4.1 Best-First Search 92 Minimize estimated cost to reach a goal: Greedy search 93 Minimizing the total path cost: A* search 96 4.2 Heuristic Functions 101 The effect of heuristic accuracy on performance 102 Inventing heuristic functions 103 Heuristics for constraint satisfaction problems 104 4.3 Memory Bounded Search 106 Contents _xv Iterative deepening A* search (IDA*) 106 SMA* search 107 4.4 Iterative Improvement Algorithms 1 Hill-climbing search 1 Simulated annealing 1 Applications in constraint satisfaction problems 1 4.5 Summary 115 Bibliographical and Historical Notes 115 Exercises 118 Game Playing 122 5.1 Introduction: Games as Search Problems 122 5.2 Perfect Decisions in Two-Person Games 123 5.3 Imperfect Decisions 126 Evaluation functions 127 Cutting off search 129 5.4 Alpha-Beta Pruning 129 Effectiveness of alpha-beta pruning 131 5.5 Games That Include an Element of Chance 133 Position evaluation in games with chance nodes 135 Complexity of expectiminimax 135 5.6 State-of-the-Art Game Programs 136 Chess 137 Checkers or Draughts 138 Othello 138 Backgammon 139 Go 139 5.7 Discussion 139 5.8 Summary 141 Bibliographical and Historical Notes 141 Exercises 145 III Knowledge and reasoning Agents that Reason Logically 6.1 A Knowledge-Based Agent 6.2 The Wumpus World Environment Specifying the environment Acting and reasoning in the wumpus world 6.3 Representation, Reasoning, and Logic Representation 149 151 151 153 154 155 157 160 Inference 163 Logics 165 6.4 Prepositional Logic: A Very Simple Logic 166 918 Index knowledge precondition, 247 knowledge representation, 5, 15, 18, 23,157,257 analogical, 213 language, 152, 178 knowledge source (KS), 770 knows that, 246 Knuth.D., 132, 143,293,685,853, 879 Knuth-Bendix algorithm, 293 KODIAK, 298 Koenderink,J., 769,579 Kohn, W., 329,879 Kohonen.T., 595,880 Koko, 653 Roller, D., 27, 87, 432,519, 520, 596, 860,877, 878, 880,893 Kolmogorov, A N., 431, 432,520, 559, 880 Kolmogorov complexity, 559 Kolodner,!, 23,646,550 Konolige, K., 260, 880 Koopmans, T., 520,880 Korf,R E., 87, 117, 880 Kotok,A., 143,880 Kowalski, R., 213, 259,292, 304, 328, 330,880 Koza,J R., 623, SSO Kripke, S A., 260,880 Krogh, A., 595,575 Krotkov,E., 778,596 Kruppa, E., 768, 880 KRYPTON, 332 Kube, P., 258, 876 Kuehner, D., 293,880 Kukich, K., 720, 880 Kulikowski, C A., 560, 901 Kumar, V., 116, 117,878,880, 881 887 Kurzweil, R., 5, 28, 881 Kyburg, H E., 432, 881 Ladkin, P., 259, 881 Ladner, R., 259, 577 Lafferty, J., 687, 862 Laird, J., 26, 316, 645, 788, 843, 881,887 Laird, P., 116,886 Lakoff, G., 258, 720, 881 lambda calculus, 329 Lambertian surface, 729,743, 771 La Mettrie, J O de, 838, 849,881 landmark, 805, 809 Langley, P., 881 Langlotz,C, 25,876 language, 651 as action, 685 analysis, 658 formal, 654 interfaces, 23, 693 model, 760 in disambiguation, 682 module, 653 "natural, 7, 161,654 origin, 653 perception, 657 problem-solving, 329 processing, 16 situated, 659, 659, 684 and thought, 162,653 translation, 21,632, 691-693,720 understanding, 19, 23 Lansky, A L., 364, 872 Laplace, P., 12,430,432,458 Larson, G., 626 laser range finder, 785 Lassez, J.-L., 328, 329, 877,881 Latin, 668 Latombe, J.-C, 390, 811, 868, 88] lattice theory, 313 Lauritzen, S., 465,596, 577, 881, 897 LAWALY, 389 Lawler.E., 116,55,? Laws, K I., 685, 876 laziness, 416, 558 leak node, 443 leaping to conclusions, 626 learning, 153,525, 823,830 action-value, 599 active, 599 architecture, 844 assessing performance, 538 Bayesian, 588, 588-589,593 belief network, 531, 588-592 blocks-world, 19 cart-pole problem, 617 checkers, 18 curve, 538 decision lists, 555-556 decision trees, 531-536,641 determinations, 634 driving, 586-587 element, 525,558 linearly separable functions, 575-577 new predicates, 638, 640 PAC, 553,560, 632 passive, 599 pronunciation, 585-586 Q, 613 rate, 576, 579,604, 613 restaurant problem, 531, 620 speedup, 527 top-down, 641-644 utility function, 599 least-commitment, 549 least-constraining-value, see heuristic least commitment, 271, 346, 374 least mean squares, 601 Le Cun, Y., 586, 595,881 Lederberg, J., 22, 94, 257, 870, 882 Lee, K.-E, 623, 770,881,900 Lee, R., 294, 864 Leech, E, 688, 872 Leech, G., 685, 597 Lefkovitz, D., 145, 882 left-corner parser, 699 legal reasoning, 29 Leibniz, W., 9, 15, 179,432, 827 Leiserson, C E., 853, 866 Lemmer, J E, 467,878 Lenat, D B., 242,258, 263, 646, 647, 828, 844, 867,882 LENGTH, 666,702,851 lens, 727 Leon, M., 692,900 Leonard, H S., 258,882 Leonard, J., 520,882 LePape, J.-R, 330, 869 Lesh,N., 411,570 Lesniewski, S., 258, 882 Lesser, V R., 770, 570 levels of description, 152 Levesque,H., 184,259,261,331, 332, 863,882, 894, 895 Levy, D N., 144,875 Lewis, D K., 50, 685, 839, 882 Lewis, R., 641,579 LEX, 552,559 explanation-based, 788 lexicon, 664,686, 698, 703-704 and failure, 403 LEG, see grammar, lexical-functional Li, M., 560, 882 liability, 848 Lieberman, L., 769, 860 to fly, 539 game-playing, 617 handwriting recognition, 586 human, 624 incremental, 529, 549,626 inductive, 558, 566, 829 knowledge-based, 628, 637, 645 knowledge in, 625—628 LIFE, 297, 329 life insurance, 479 Lifschitz, V., 363, 390,466,872, 882 lifting lemma, 286,289 Index 919 light-beam sensors, 785 Lighthill, J., 21,882 Lighthill report, 21,24 likelihood, relative, 427 likelihood weighting, 456,465 limb (in a scene), 745 limited rationality, Lin, S., 115 Linden, T A., 411,852 Lindsay, R K., 257,720,882 linear input resolution, 312 linearization, 346, 347 linear plan, see plan, linear linear resolution, 285 linear separability, 574,593 line labelling, 745 linguistics, 15-16,28 computational, 16 Linguistic String Project, 685 link (in a neural network), 567 link (of a robot), 777, 809 linked list, 329 linking curve, 800 LINKS, 347, 372,375 Linnaeus, 258 Lipkis, T., 332,894 LIPS, see logical inferences per second liquid event, see event, liquid liquids, 260 Lisp, 18, 197,329,654 Common, viii, 329,330, 855 lists, 200 literal (sentence), 167, 182 Littman, M., 520,864 lizard toasting, 627 LLA, see action, low-level Lloyd, J., 328,882 LMS, see least mean squares LMS-UPDATE, 602,602, 623 LMT (Lozano-Perez, Mason, and Taylor), 795 local encoding, 577 locality, 174,460 locally finite graph, 100 locally structured system, 441 Locke, J., Locke, W N., 720,882 locking, 312 locomotion, 777, 778 Lodge, D., 848 logic, 7, 11,151,158 autoepistemic, 466 combinatory, 329 default, 459 description, 298 dynamic, 259 first-order, 185,211 semantics, 186 syntax, 186 first-order probabilistic, 843 fuzzy, 166,417,459,463^66 higher-order, 195 inductive, 432,432 < modal, 245, 258,260 nonmonotonic, 321,459 notation, prepositional, 151,165 limitations, 176 semantics, 168 syntax, 166 temporal, 165,258, 259 terminological, 298 logical connective, 16, 165, 189 logical inferences per second, 307 logical level, 153 logical omniscience, 246 logical piano, 292 logical positivism, 10 logicism, 7, 16 logic programming, 293, 304-310, 328, 343 constraint, 308, 328 inductive (ILP), 628,636-645 language, 297 logic sampling, 455,465, 515 Logic Theorist, 292 Logic Theorist (LT), 17 logic variables, 306,308 LOGIN, 328,329 long-distance dependency, 710 long-term memory, 316 Longuet-Higgins, H., 769, 882 LOOKUP, 38, 321 lookup table, 38, 572 LOOM, 298 lottery, 473 standard, 478 love, 823 Lovejoy.W., 520,883 Lovelace, A., 15,823 Loveland.D., 293,294,553 Lowe, D., 769,88.3 Lowenheim, L., 212, 883 Lowerre, B., 770,883 Lowrance, J., 466,893 Lowry, M R., 330,883 Loyd, S., 86, 883 Lozano-Perez, T., 795, 883 LT, 17 LT (Logic Theorist), 17 Luby, M., 465,867 Lucas, J., 824, 883 Luger.G R, 5,883 LUNAR, 23,693,720 Lunokhod robots, 775 Lusk, E., 294, 902 M M? (model), 499 M (model), 605 MacHack 6, 143 machine evolution, 21 machine learning, 5, 7,463, 687, 829 machine translation, see language, translation Machover.M., 213,867 Mackworth, A., 749,769, 853 Macpherson, P., 466, 899 macro-operator, 787 macrop, 389 Maes, P., 36, 553 Magerman, D., 687, 562,583 Mahajan, S., 623,881 Mahanti,A., 117,883 Mahaviracarya, 431 Maher, M., 328, 881 MAJORITY-VALUE, 537 majority function, 533,573 MAKE-ACTION-QUERY, 152, 177, 202, 663 MAKE-ACTION-SENTENCE, 152, 202, 338, 395,408 MAKE-GOAL-QUERY, 337,338,408 MAKE-MINIMAL-PLAN, 356 MAKE-NODE, 73,107,110,112, 113,666 MAKE-PERCEPT-SENTENCE, 152, 177,202,338,395,402, 408,663 MAKE-PLAN, 399,408 MAKE-QUEUE, 72,73,110 Malik, J., 27,520, 737, 743, 749, 769, 877,880, 883 Manhattan, see heuristic, Manhattan MANIAC, 587 Manin, Y., 293,553 manipulation, 254,725,777 manipulator, 780 Mann.W.C., 719,720, 553 Manna, Z., 213,293, 330, 553 manufacturing, 390 MAP, see maximum a posteriori map-coloring problem, 91, 105 mapping, 34 ideal, 34 Marchand, M., 595, 883 Marin, J., 559, 877 Mark MI/HI, 14 Markov, A A., 500,770, 553 Markov blanket, 465 Markov chain, 514 Markov decision problem, 411,500 920 Index partially observable, 500 Markov decision problems, 520 Markov model, 762 Markov property, 500,509,762, 765 Marr, D., 769,883 Marriott, K., 328, 881 Marsland, A T., 144, 884 Martelli, A., 116, 143, 328,884 Martin, C, 28, 884 Martin, ]., 720, 884 Martin, P., 258, 694, 720, 874, 876 MARVEL, 26 Maslov,S., 293,884 Mason, M., 411, 795, 883,884 mass noun, 242 mass spectrometer, 22 MATCH, 666 match phase, 314 material advantage, 127 material handling, 774 materialism, 9, 819 eliminative, 840 material value, 127 Mates, B., 119,884 mathematical induction schema, 288 mathematical objection, 824, 826 mathematics, 11-12 Matheson, J., 484,494, 576,886 Mauchly,J., 14 MAUT, see multiattribute utility theory MAX, 110,132 MAX-VALUE, 130-132,132 maximum a posteriori, 588 maximum expected utility, 472,493, 502 maximum likelihood, 589 Maxwell,!., 458,720,554 Mayer.A., 116,574 Mays, E., 325.554 McAllester, D., 143, 312 330, 332, 364, 554 MCC, 24 McCarthy, J., 17, 18,50 179,212, 213,230,257,259,329, 459, 820, 832, 554 McCartney, R D., 330, 883 McCawleyJ D., 685,555 McClelland, J L 24, 893 McConnell-Ginet, S., 28, 685.565 McCorduck, P., 810,555 McCulloch, W S., 16, 19,563, 570, 594, 555 McCune, W., 310, 330, 555 McDermott, D., 5, 213, 317, 328, 330,331,390,411,459, 565, 555, 596, 595 McDermott, J., 24, 316, 555 McDonald, D., 707 McDonald, R., 185 McGuinness, D., 324, 332, 863 MDL, see minimum description length MDP, see Markov decision problem Mead, C., 595,555 meal, 251 meaning, 161 means-ends analysis, 10,10 mean square error, 623 measure, 231, 386 measure fluent, 386 medical diagnosis, 22, 23, 27, 465 MeehanJ., 328,565 Meet, 239 meganode, 453 Megarian school, 179,258, 291 Megiddo, N., 847,555 Melcuk, I., 687,555 Mellish, C., 330, 865 micromort, 480,494,497 microworld, 19,19, 21, 827 Middleton, B., 444,456, 890 Miles, E, 596, 885 Mill, J S., 10, 546, 555, 556 Miller, A., 494,556 Miller, D., 389,565 Miller, G., 595, 704, 556 Milne, A A., 218, 556 MEMBER, 307,307,322 MlNIMAL-CONSISTENT-DET, 635, MEMBER?, 322 635 minimax, see search, minimax MEMBERSHIPS, 321,322 memoization, 87,452, 629 memory requirements, 75, 77, 79 meningitis, 426^127,434, 435 mental model, in disambiguation, 682 mental objects, 243-247 mental states, 819 Mercer, R., 687, S62 mereology, 258 Merkhofer, M., 494, 556 MERLIN, 331 Mero, L., 116,555 meta-comment, 718 Meta-DENDRAL, 552, 559,641 METALOG, 330 metamer, 730 metaphor, 715, 720, 822 metaphysics, 10 metareasoning, 140,309, 364 decision-theoretic, 844 metonymy, 714, 720 Metropolis, N., 117,555 Metropolis algorithm, 117 MEU, see maximum expected utility Mezard, M., 573, 595, 555 MGSS*, 143 MGU, see unifier, most general Michalski, R S., 560, 555 Michaylov.S., 329,577 Michie, D., 28, 70, 86, 115,117, 144,221,539,560,618, 622,810,569,555,594 MICRO-PLANNER, 330 microelectronics industry, 774 MIN, 107, 132 min-conflicts, see heuristic, min-conflicts, 116 MIN-VALUE, 130, 132,132 mind, 5, 817, 838 conscious, 818 dualistic view, 838 and mysticism, 565 philosophy of, 817, 838, 840 as physical system, 8, and symbolism, 180 theory of, MINIMAX-DECISION, 126,147 MINIMAX-VALUE, 126,126, 135, 147 minimax decision, 124 minimization (logical), 236 minimum description length, 560 minimum spanning tree, 116,119 Minker,J., 328,572,556 Minsky.M L., 16,18,21,23,24, 28,331,465,577,578,594, 595, 827, 556 Minton.S., 116,645,556 missing attribute values, 543 missionaries and cannibals, 67, 86, 88 MIT, 17, 18,781,810 Mitchell, D., 184,595 Mitchell, T., 552, 559,560, 645, 788,812,843,565,564, 555, 556 Mitchell, T M., 645 MIT1, 777 ML, see maximum likelihood ML (programming language), 329 mobile robots, 520 mobot, 775 modal logic, see logic, modal model causal, 443 (in logic), 170 (in representation), 13 theory, 212 trees, 616 model-based reasoning, 209 INDEX 921 model-based recognition, 785 modification operator, 346 modularity, 210 Modus Ponens, 172,245,285, 290, 291,294,819 Generalized, 269,269-270 MOLGEN, 390 monkey and bananas, 366 monotonicity (of a heuristic), 97, 116 monotonicity (of a logic), 173, 321, 459 monotonicity (of preferences), 474 Montague, R., 258, 259, 685, 878, 886, 899 Montanari, U., 116, 143,328, 884 Monty Python, 159 Mooney.R., 645,868 Moore, A., 623,886 Moore,J., 331, 886 Moore, J.S., 293,313,328, 330, 826, 863 Moore, R., 132,143,260,261,466, 876, 879, 887 morality, 817 Moravec, H., 835, 849, 887 More,!., 17 Morgan, J., 720, 866 Morgenstern, L., 260, 887 Morgenstern,O., 12, 142,493,847, 900 morphology analysis, 695, 703 derivational, 703 inflectional, 703 Morris, P., 467, 873 Morrison, E., 142,887 Morrison, P., 142, 887 Mosher.R., 776 most-constrained-variable, see heuristic most-constraining-variable, see heuristic most general unifier, see unifier, MPI, see mutual preferential independence MRS, 297, 309,313,330,457 MST (minimum spanning tree), 119 Muggleton, S., 466, 641,646, 860, 879, 887 neat, 25 needle in a haystack, 160 negated literals, 312 negation, 167 jnultiagent domain, 244 multiattribute utility theory, 480, 480, 494 multilayer network, see neural network, multilayer multiple-state problem, see problem multiple inheritance, 320 multiply connected network, 453 negligence, 848 Mundy.J., 769,887,893 MUNIN, 465 Murphy's Law, 59, 87 music, 15 mutation, 21, 619,620 mutually utility-independent, 484 mutual preferential independence, 483 MYCIN, 23-24,461,466 myopic policy, 490 Myrhaug.B., 331,862, 867 mysticism, 565 N n-armed bandit, 611 Nadal,J.-P., 573,595,885 Nagel, T., 839, 887 Nalwa,V S., 13,770,887 NAME, 321 naming policy, 660 narrow content, 821 NASA, 390,693,776 NASL, 411 Nasr, R., 328,859 natural deduction, 291 natural kind, 232, 319 natural language, see language, natural Natural Language Inc., 694 natural language processing (NLP), see language negation as failure, 304, 343 negative example, 534 Neisser, U., 594,895 Nelson, H., 144, 877 NERF, 572 NETL, 298,331 NETtalk, 585 Netto, E., 86, 887 NEURAL-NETWORK-LEARNING, 576, 577,580 NEURAL-NETWORK-OUTPUT, 577, 597 neural computation, see neural network neural network, 16, 19, 24, 128, 139,530,563,563,593,829 vs belief network, 592 efficiency, 583 expressiveness, 16, 583 feed-forward, 570, 593 hardware, 16 learning, 16 multilayer, 21,571, 593 and noise, 584 nonlinearity in, 567 recurrent, 570 second best, 585 neurobiology,565, 769 Neurogammon, 617 neuron, 16,452,563,564,819.833, 835 NEW-CLAUSE, 643 NEW-LITERALS, 642-644 NEW-VARIABLE, 307 Newborn, M., 117,889 Newell, A., 6, 10,17,26,86,94, 115, 143,179,292,316, 329,331,645, 843, 587, 886-888,896 898 Newton, I., 3, 509 most general Mostow,J., 116,118,587 natural science, 654 NEXT-SUCCESSOR, 110 Motet, S., 27, 879 natural stupidity, 317 Nau, D., 870 Neyman, A., 847, 888 Nicholson, A., 520, 868, 888 Nicholson, A E., 520 Nievergelt, J., 145, 879 Nilsson, N J., 28, 87, 115,116,118, motion, 735-737 compliant, 803 guarded, 803 motion parallax, 737, 768 motion planning, 796-808,811 complexity, 811 motor subsystem, 410 Motzkin, T., 594, 859,887 Mourelatos, A P., 259,887 Moussouris, J., 144,887 Nau, D S., 116,143,389,390,878, 887 Naur, P., 685,887 Navathe, S., 325, 867 navigation, 252-253,725, 796-808 landmark-based, 796 NavLab, 586 NAVTO, 787 Neal, R., 596,887 Neapolitan, R E., 467, 887 143, 144,213,286,293, 363,411,594,645,810, U, 870,872, 874,888, 900 Nim, 142 Nitta, K., 329, 888 Nixon, R., 320, 682,714 Nixon diamond, 320 922 Index NL-MENU, 720 NLP, see natural language processing NOAH, 330, 363,411 Nobel prize, 50 node, search, 71 noise, 535, 542-543,552 563, 584, 588,636,731.795 noisy-OR, 443 nominal compounds, 706, 721 nominalist, 258 nominative case, see subjective noncomputability, 11 nondeterminism, 855-856 nondeterministic, see environment, nondeterministic NONDETERMINISTIC-CHART-PARSE, 699 nondeterministic algorithm, 855 nonepisodic, see environment nonepisodic nonholonomic, see robot, nonholonomic NONLIN, 363 NONLIN+, 389, 390 nonlinear plan, see plan, nonlinear nonmonotonicity, 321,459 nonmonotonic logic, see logic, nonmonotonic Nono, 267 nonterminal symbol, 655, 656, 854 normalization, 428 normalizing constant, 451 Normann, R A., 832, 864 normative theory, 479 North, O., 267 North,!., 21, 871 Norvig, P., 328, 330, 688, 770, 878, 888 notation arithmetic, logical, notational variations for FOL, 196 noun phrase, 655, 709 Nowatzyk.A., 144, 876 Nowick, S., 313,888 NP, see noun phrase NP-completeness, 12, 21, 852, 853 nuclear power, 468, 838 NUMBER-OF-SUBKINDS, 322 number theory, 647 Nunberg,G., 720, 888 Nussbaum,M.C., 888 Nygaard.K., 331,862,867 O O() notation, 852 cP (configuration space obstacle), 791 O'Brien, S., 720 891 O'Keefe, R., 330 888 abstract, 372 primitive, 372 O-PLAN, 369-371.387,390 operator expansion, see planning, hierarchical operator reduction, see planning, object, 165,185,188 ^composite, 234 OPERATORS, 60,73,126 object-oriented programming, 15 object creation, 384 objective case, 668 objectivism, 430 object recognition, 725, 751-755 observation sentences, 10 obstacle avoidance, 808 OCCUR-CHECK, 303 occur-check,303,305 Ochiai, T., 618, 871 Ockham's razor, 534 558, 559, 589, 626,644 Ockham,W., 534,559 Ockham algorithm, 560 octant, 747 odometry, 783 Oetzel, R M., 596, 900 offline cost, 61 Ogasawara, G., 27,520, 877 880 Oglesby.E, 313, 330,574 hierarchical operator schema, 344 OPS-5,298,314,316 optical character recognition, 657 optical flow, 735, 750, 769 optimality (of a search algorithm), 73,85 optimally efficient algorithm, 99 optimal solution, 76 optimistic prior, 610 optimization problem, 619 optimizer, peephole, 379 OPTIMUM-AIV, 367,369, 390 Opus, 320 OR, see operations research Or-Introduction, 172 oracle, 856 orderability, 474 ordering constraints, 350 ORDERINGS, 347, 372,374 ordinal utility, see utility, ordinal Olalainty, B., 390,871 Organon, 179,257 Olawsky.D., 411,888 Olesen, K., 465, 860, 877, 888 Ortony, A., 720,888 Oliver, R., 494 888 Olson, C., 754, 888 Olum, P., 769, 872 omniscience, 32, 558 logical, 246 Osherson, D N., 559, 889 Othello, 138,623 OTTER, 297,310, 311,311,313, 330,333 outcome of a lottery, 473 output unit, 571 one-shot decision, see decision, Overbeek, R., 294, 902 one-shot online algorithm, 796, 806, 806-809 overfitting, 542, 542-543,572, 588 overgeneration, 668 online cost, 61 overshooting, 577 online navigation, 813 Ono,N., 618,57; OWL, 298 ONTIC, 312, 330 ontological commitment, 165, 185, 417,459 P (probability), 420 P (probability vector), 421 ontological engineering, 217, 222 ontological promiscuity, 258 PAC, see probably approximately correct packed forest, 697, 703 ontology, 222, 257 Op (STRIPS Operator), 344 OP, 299, 303 opacity, 244 open-coding, 306 open class, 664 operationality, 632 operationality and generality, 632 operations research, 86, 87, 116, 117,367,498,500 OPERATOR, 72 operator, 60, 85, 123 packed tree, 723 Paek, E., 595,870 PAGE description, 36, 37, 248 Paige, R., 330, 889 Palay, A J., 143,889 Palmer, R G., 595, 875 Palmieri, G., 594, 872,889 Pandemonium, 594 Pang, C., 87, 868 Panini, 15, 685 Pao, C., 26, 903 Index 923 Papadimitriou C., 749,847, 879, 889 Papert, S., 21, 24,577, 578,594, 595, 886 Peano.G., 212, 316,889 Pearl, J., 25, 97, 99, 116-117,143, 435,437,465,467,497 596, 868, 873, 879, 889,895 PARADISE, 140 PEDESTAL, 390 paradoxes, 259 parallel distributed processing, see Pednault,E P., 390,889 neural network Parallel Inference Machine, 308 parallelism, 566 AND-, 308 OR-, 308 parallel lines, 726 parallel search, 117, 117 paramodulation, 284,293 PARENT-NODE, 72 parent node, 72 parity function, 533 Parker, D., 595, 889 PARLOG, 329 Parrod, Y., 390, 859 PARSE, 662,663 parse forest, 664 parse tree, 658, 664,701 parsing, 658,658, 664 chart, 697 partial derivative, 580 partial order, see planning, partial-order partial plan, see plan, partial partition, 231, 795 part of, 233 part of speech, 658 Partridge, D., 29, 889 parts, 252 Pascal's wager, 432, 493 Pascal, B., 12, 14,431 Pascaline, 15 Pasero, R., 328, 866 PASSIVE-RL-AGENT, 602,605,607 passive learning, 599 passives, 687 Paterson, M., 889 path, 60, 85 PATH-COST, 72 PATH-COST-RlNCTION, 60 path cost, 60, 61,85 PATHFINDER, 457,458,465 pathmax, 98 Patil, R., 332,702,720, 865,869 Patrick, B., 117,889 pattern matching, 330 Paul, R P., 811,889 paying, 255 payoff function, 124,418 PCFG, see grammar, context-free POP, see parallel distributed processing PEGASUS, 26 Peirce,C S., 212, 316.323.685, 889 pen, 713 Penberthy, J., 390, 861,889 Peng, J., 623, 889 Pengi, 411 Penrose, R., 825, 889 Pentland, A., 36, 883 Peot, M., 411,465, 889,895 PERCEPT, 48 percept, 39 perception, 30 perceptron, 19, 571,573, 593, 594 convergence theorem, 20 learning rule, 576, 593 representational power, 21, 596 percept sequence, 33 Pereira, E, 304, 306, 685,688, 693, 694, 720,874, 889, 890, 900 Pereira, L., 306, 900 PERFORMANCE-ELEMENT, 608, 609 PERFORMANCE-FN, 48 performance element, 525, 526, 558,562,766 performance measure, 32, 40 47, 50,416,472 Perrault, C., 720, 866 perspective 767 perspective projection, 726, 735 PERT, 367-369 Peters, S., 685,869 Peterson, C., 595, 890 Petrie, T., 770, 867 phenomenology, 828 Philips, A., 116, 886 Phillips, M., 26, 903 Phillips, S., 87, 878 Philo, 179 philosophy, 3, 8-10, 817-841 European, phone (speech), 757 phoneme, 585 phone number, 246 phonetic alphabet, 758 photogrammetry, 768 photometry, 729 photosensitive spot, 749 phrase, 655 phrase structure, 655, 685 physicalism, 819,839 Picasso, P., 834 pick (choice point), 856 Pickwick, Mr., 831 piecewise continuity 740 pigeons, 13 Pillsbury, 696 PIM, see Parallel Inference Machine ping-pong, 29, 598 pinhole camera, 725 Pinker, S., 687,890 Pisa, tower of, 526 pit, bottomless, 153 Pitt, L., 560,862 Pitts, W., 16,19, 563, 570, 594, 885 pixel, 727 place, 236 Place, U., 839, 890 plan, 347 canned, 407 complete, 349 conditional, 410, 806 consistent, 349, 349 fully instantiated, 346 initial, 347 linear, 363 noninterleaved, 363 nonlinear, 363 parameterized, 398 partial, 345 representation, 346 PLAN-ERS1, 390 PLANEX, 411,787 Plankalkul, 14 PLANNER, 23,330,402 planning, 42, 140, 211,342, 539 abstraction, 389 adaptive, 389 ADL formalism, 390 assembly, 792 blocks world, 19 bounded-error, 796 case-based, 389 conditional, 392, 393-398,407, 412,415,500 contingency, 392 andDDNs, 519 deferred, 393 and execution, 403-406 fine-motion, 802, 809 formalization, 25 hierarchical, 371-380,389 hierarchical decomposition, 374 history, 363 as logical inference, 341 menu, 249-252 multi-agent, 390 partial-order, 337, 346 355-356, 390,407 progression, 345, 365 924 Index reactive, 411, 411 regression, 345, 356, 363 route, 18 and scheduling, 369 search spaces, 345-346 situation calculus, 341-342 situation space, 345 speech acts, 654 symbolic, 788 total order, 346 under uncertainty, 795, 843 planning agent, see agent, planning planning variable, 400 plan recognition, 654,718 plasticity, 564 Plato, 8, 178,827,838 Plotkin, G D., 646, 890 ply, 124 Pnueli, A., 259,890 Podelski, A., 329, 859 poetry, 4, 682 Pohl, I., 87, 116,890 poker hands, 433 Poland, 240 Polguere, A., 687, 885 policy, 411,498,500, 517,519, 806 optimal, 500 POLICY-ITERATION, 506,608 policy iteration, 505,505-506,520, 603, 843 policy loss, 505 Polifroni, J., 26, 903 Pollack, M., 720, 866 Pollard, C, 686, 890 Polya, G., 94,890 polytree, 448,464 POMDP, see Markov decision problem Pomerleau, D A., 26, 586, 587,750, 877, 890 POMPD (partially observable Markov decision problem), 520 Pooh, Winnie the, 218 POP, 355, 356, 356, 357, 358, 362, 364-367,369,374,384, 391,398,404,406,412 POP-DUNC, 381, 384, 385, 390, 401 POPLOG, 330 Popper, K R., 432, 559, 836, 890 Port-Royal Logic, 471, 493 Portuguese, 627 pose, 734,752 positive example, 534 positivism, logical, 10 possibility theory, 466,466 possible threat, 354,357, 357 possible world, 260,795 Post, E L., 179,890 posterior probability, see probability, conditional POSTSCRIPT, 706, 707 Prade, H., 466, 869 Pradhan, M., 444, 456,890 pragmatic interpretation, see interpretation, pragmatic pragmatics, 658 Prata, A., 595, 870 Pratt, V R., 259, 890 Prawitz, D., 291,293,890 pre-editing, 692 precompilation, 270 precondition, 344 disjunctive, 383 universally quantified, 383 PRECONDITIONS, 402 predecessor, 80 predicate, 165,674 predicate calculus, see logic, first-order predicate symbol, 187, 211 PREDICT, 723 predicting the future, 531, 553 prediction-estimation process, 515 prediction phase, 509 PREDICTOR, 698-700,702,702 predictor, 698 preference, 418, 473,474, 483 lexicographic, 496 monotonic, 476 preference independence, 483 preferentially independent, 496 premise, 167 prenex form, 292 preposition, 664 president, 240 Presley, E., 240 PRESS, 330, 331 Price Waterhouse, 369 Prieditis, A., 103, 116,118,887,890 primary colors, 730 Princeton, 17,704 Principia Mathematica, 17 PRINT, 856 Prinz, D., 142, 890 Prior, A N., 258,890 prioritized sweeping, 607, 623 priority queue, 623 prior knowledge, 625,627, 636, 645 prior probability, see probability, prior prismatic motion, 781 Prisoner's Dilemma, 847 prisoners, three, 435 probabilistic model, 425 probabilistic network, see belief network adaptive, 590 probabilistic projection, 515 probabilistic sensor models, 520 probability, 12,23,25, 127 alternatives to, 458 assessment, 430 axioms of, 422-423 conditional, 418,421, 421-422, 431,440 conjunctive, 440 distribution, 139,421,445 history, 433 judgments, 443 of sun rising, 430 prior, 420,420-421,431 theory, 417, 493 probably approximately correct, 553, 556, 560 PROBLEM,60 problem, 60, 85 airport-siting, 496 assembly sequencing, 70 bandit, 623 contingency, 59, 123 cryptarithmetic, 65, 91 datatype, 60 8-queens, 64, 83 8-puzzle, 101,103, 115 8-queens, 86, 89 exploration, 59 halting, 277, 824 inherently hard, 852-853 intrinsically difficult, 106 map-coloring, 91, 105 missionaries and cannibals, 67 monkey and bananas, 366 multiple-state, 58, 66 real-world, 63 relaxed, 103 robot navigation, 69 route-finding, 93 shoes-and-sock, 347 shopping, 340 single-state, 58, 60 toy, 63 travelling salesperson, 69 VLSI layout, 69, 114 problem-solving agent, see agent, problem-solving problem formulation, 56, 57 problem generator, 526, 562 problem solving, 22 complexity, 342 vs planning, 338 procedural attachment, 323 PROCESS, 311,311 Index 925 process, 237 production, 40, 854 production system, 297, 314 product rule, 421 Putnam, H., 50, 286, 292, 432, 839, 867, 891 Pylyshyn.Z W., 839,