人工智能（英文版）

     Contents
   Preface
   1 Introduction
    1.1 What Is AI?
    1.2 Approaches to Artificial Intelligence
    1.5 BriefHistoryofAI
    1.4 Plan of the Book
    1.5 Additional Readings and Discussion Exercises
   I Reactive Machines
   2 Stimulus-Response Agents
    2.1 Perception and Action
    2.1.1 Perception
    2.1.2 Action
    2.1.5 Boolean AIgebra
    2.1.4 Classes and Forms ofBoolean Functions
    2.2 Representing and Implementing Artion Functions
    2.2.1 Production Systems
    2.2.2 Networks
    2.2.3 The Subsumption Architecture
    2.3 Additional Readings and Discussion Exercises
   Neural Networks
    3.1 Introduction
    3.2 Training Single TLUs
    3.2.1 TLU Geometry
    3.2.2 Augmented Vectors
    3.2.3 Gradient Descent Methods
    3.2.4 The Widrow-Hoff Procedure
    3.2.5 The Generalized Delta Procedure
    3.2.6 The Error-Correction Procedure
    3.3 Neural Networks
    3.3.1 Motivation
    3.3.2 Notation
    3.3.5 The Backpropagation Method
    3.3.4 Computing Weight Changes in the Final Layer
    3.3.5 Computing Changes to the Weights in Intermediate Layers
    3.4 Generalization, Accuracy, and Overfitting
    3.5 Additional Readings and Discussion Exercises
   Machine Evolution
    4.1 Evolutionary Computation
    4.2 Genetic Programming
    4.2.1 Program Representation in GP
    4.2.2 TheGPProcess
    4.2.3 Evolving a Wall-Following Robot
    4.5 Additional Readings and Discussion Exercises
   State Machines
    5.1 Representmg the Environment by Feature Vectors
    5.2 Elman Networks
    5.5 Iconic Representations
    5.4 BIackboard Systems
    5.5 Additional Readings and Discussion Exercises
   B Robot Vision
    6.1 Introduction
    6.2 Steering an Automobile
    6.3 Two Stages of Robot Vision
    6.4 Image Processing
    6.4.1 Averaging
    6.4.2 Edge Enhancement
    6.4.3 Combining Edge Enhancement with Averaging
    6.4.4 Region Finding
    6.4.5 Using Image Attributes Other Than Intensity
    6.5 Scene Analysis
    6.5.1 Interpreting Lines and Curves in the Image
    6.5.2 Model-Based Vision
    6.6 Stereo Vision and Depth Information
    6.7 Additional Readings and Discussion Exercises
   II Search in State Spaces
   Agents That Plan
    7.1 Memory Versus Computation
    7.2, State-Space Graphs
    7.5 Searching Explicit State Spaces
    7.4 Feature-Based State Spaces
    7.5 Graph Notation
    7.6 Additional Readings and Discussion Exercises
   Umnformed Search
    8.1 Formulating the State Space
    8.2 Components of Implicit State-Space Graphs
    8.5 Breadth-First Search
    8.4 Depth-First or Backtracking Search
    8.5 Iterative Deepening
    8.6 Additional Readings and Discussion Exercises
   Heuristic Search
    9.1 Using Evaluation Functions
    9.2 A General Graph-Searching Algorithm
    9.2.1 Algorithm A
    9.2.2 Admissibility of A
    9.2.3 The Consistency (or Monotone) Condition
    9.2.4 Iterative-Deepening A
    9.2.5 Recursive Best-First Search
    9.5 Heuristic Functions and Search Efficiency
    9.4 Additional Readings and Discussion
   Exercises
    10Planning, Acting, and Leaming
    l0.l The Sense/Plan/Act Cycle
    10.2 Approximate Search
    10.2.1 Island-Driven Search
    l0.2.2 Hierarchical Search
    l0.2.3 Limited-Horizon Search
    l0.2.4 Cycles
    l0.2.5 Building Reactive Procedures
    l0.3 Leaming Heuristic Functions
    l0.3.l Explicit Graphs
    10.3.2 Implicit Graphs
    l0.4 OA Rewards Instead of Goals
    l0.5 Additional Readings and Discussion
   Exercises
    11Altenuative Search Fomulations and Applications
    ll.l Assignment Problems
    ll.2 Constructive Methods
    11.3 Heuristic Repair
    11.4 Function Optimization
   Exercises
    12Adversarial Search
    12.1 Two-Agent Games
    12.2 The Minimax Procedure
    12.3 The Alpha-Beta Procedure
    12.4 The Search Effidency of the Alpha-Beta Procedun
    l2.5 Other Important Matters
    12.6 Games of Chance
    12.7 Learning Evaluation Functions
    l2.8 Additional Readings and Discussion
   Exercises
   III Knowledge Representation and Reasoning
   The Propositional Calculus
    13.1 Using Constraints on Feature Values
    13.2 The Language
    13.3 Rules of Inference
    l3.4 Definition of Proof
    13.5 Semantics
    13.5.1 Interpretations
    13.5.2 The Propositional Truth Table
    l3.5.5 Satisfiability and Models
    l3.5.4 Validity
    l3.5.5 Equivalence
    l3.5.6 Entailment
    13.6 Soundness ahd Completeness
    13.7 The PSAT Problem
    13.8 Other Important Topics
    13.8.1 Language Distinctions
    l3.8.2 Metatheorems
    l3.8.3 Associative Laws
    13.8.4 Distributive Laws
   Exercises
    14Resolution in the Propositional Calculus
    14.1 A New Rule of Inference: Resolution
    14.1.1 Clauses as wffs
    14.1.2 Resolution on Clauses
    14.1.3 Soundness of Resolution
    14.2 Converting Arbitrary wffs to Conjunctions of CIauses
    14.3 Resolution Refutations
    14.4 Resolution Refutation Search Strategies
    14.4.1 Ordering Strategies
    14.4.2 Refinement Strategies
    14.5 Hom Clauses
   Exercises
    15 The Predicate Calculus
    15.1 Motivation
    15.2 The Language and Its Syntax
    15.3 Semantics
    15.3.1 Worlds
    15.3.2 Interpretations
    15.3.3 Models and Related Notions
    15.3.4 Knowledge
    15.4 Quantification
    15.5 Semantics of Quantifiers
    15.5.1 Universal Quantifiers
    15.5.2 Existential Quantifiers
    15.5.3 Useful Equivalences
    15.5.4 Rules of Inference
    15.6 Predicate Calculus as a Language for Representing Knowledge
    15.6.1 Conceptualizations
    15.6.2 Examples
    15.7 Additional Readings and Discussion
   Exerdses
    16 Resolution in the Predicate Calculus
    16.1 Unification
    16.2 Predicate-Calculus Resolution
    16.5 Completeness and Soundness
    16.4 Converting Arbitrary wffs to Clause Form
    16.5 Using Resolution to Prove Theorems
    16.6 Answer Extraction
    16.7 The Equality Predicate
    16.8 Additional Readings and Discussion
   Exercises
   H Knowledge-Based Systems
    17.1 Confronting the Real World
    17.2 Reasoning Using Hom Clauses
    17.3 Maintenance in Dynamic Knowledge Bases
    17.4 Rule-Based Expert Systems
    17.5 Rule Learning
    17.5.1 Leaming Propositional Calculus Rules
    17.5.2 Leaming First-Order Logic Rules
    17.5.3 Explanation-Based Generalization
    17.6 Additional Readings and Discussion
   Exercises
   Representing Commonsense Knowledge
    18.1 The Commonsense World
    18.1.1 What Is Commonsense Knowledge?
    18.1.2 Difficulties in Representing Commonsense Knowledge
    18.1.3 The Importance of Commonsense Knowledge
    18.1.4 Research Areas
    18.2 Time
    18.3 Knowledge Representation by Networks
    18.3.1 Taxonomic Knowledge
    18.3.2 Semantic Networks
    18.3.3 Nonmonotonic Reasoning in Semantic Networks
    18.3.4 Frames
    18.4 Additional Readings and Discussion
    Exercises
    19 Reasoning with Uncertain Information
    19.1 Review of Probability Theory
    19.1.1 Fundamental Ideas
    19.1.2 Conditional Probabilities
    19.2 Probabilistic Inference
    19.2.1 A General Method
    19.2.2 Conditional Independence
    19.3 Bayes Networks
    19.4 Pattems of Inference in Bayes Networks
    19.5 Uncertain Evidence
    19.6 D-Separation
    19.7 Probabilistic Inference in Polytrees
    19.7.1 Evidence Above
    19.7.2 Evidence Below
    19.7.3 Evidence Above and Below
    l9.7.4 A Numerical Example
    19.8 Additional Readings and Discussion
   Exercises
    20 Learning and Acting with Bayes Nets
    20.1 Leaming Bayes Nets
    20.1.1 Known Network Structure
    20.1.2 Learning Network Structure
    20.2 Probabilistic Inference and Action
    20.2.1 The General Setting
    20.2.2 An Extended Example
    20.2.3 Generalizing the Example
    20.3 Additional Readings and Discussion
   Exercises
   IV Planning Methods Based on
   Logic
    21 The Situation Calculus
    21.1 Reasoning about States and Actions
    21.2 Some Difficulties
    21.2.1 Frame Axioms
    21.2.2 Qualifications
    21.2.3 Ramifications
    21.3 Generating Plans
    21.4 Additional Readings and Discussion
   Exercises
   Planning
    22.1 STRlPS Planning Systems
    22.1.1 Describing States and Goals
    22.1.2 Forward Search Methods
    22.1.3 Recursive STRlPS
    22.1.4 Plans with Run-Time Conditionals
    22.1.5 The Sussman Anomaly
    22.1.6 Backward Search Methods
    22.2 Plan Spaces and Partial-Order Planning
    22.3 Hierarchical Planning
    22.3.1 ABSTRlPS
    22.3.2 Combining Hierarchical and Partial-Order Planning
    22.4 Leaming Plans
    22.5 Additional Readings and Discussion
   Exercises
   V Communication and Integration
    23 Multiple Agents
    23.1 Interacting Agents
    23.2 Models of Other Agents
    23.2.1 Varieties of Models
    23.2.2 Simulation Strategies
    23.2.3 Simulated Databases
    23.2.4 The intentional Stance
    23.3 A Modal Logic of Knowledge
    23.3.1 Modal Operators
    23.3.2 Knowledge Axioms
    25.3.3 Reasoning about Other Agents' Knowledge
    25.3.4 Predicting Actions of Other Agents
    23.4 Additional Readings and Discussion
   Exercises
   Communication among Agents
    24.1 Speech Acts
    24.1.1 Planning Speech Acts
    24.1.2 Implementing Speech Acts
    24.2 Understanding Language Strings
    24.2.1 Phrase-Structure Grammars
    24.2.2 Semantic Analysis
    24.2.3 Expanding the Grammar
    24.3 Efficient Communication
    24.3.1 UseofContext
    24.3.2 Use of Knowledge to Resolve Ambiguities
    24.4 Natural Language Processing
    24.5 Additional Readings and Discussion
   Exercises
   Agent Architectures
    25.1 Three-Level Architertures
    25.2 Goal Arbitration
    25.3 The Triple-Tower Architecture
    25.4 Bootstrapping
    25.5 Additional Readings and Discussion
   Exercises
   Bibliography
   Index