Artificial_Intelligence：A_Modern_Approach，4th_Edition.pdf

【核心代码】

Contents
I Artificial Intelligence
1 Introduction 1
1.1 What Is AI? 1
1.2 The Foundations of Artificial Intelligence 5
1.3 The History of Artificial Intelligence 17
1.4 The State of the Art 27
1.5 Risks and Benefits of AI 31
Summary 34
Bibliographical and Historical Notes 35
2 Intelligent Agents 36
2.1 Agents and Environments 36
2.2 Good Behavior: The Concept of Rationality 39
2.3 The Nature of Environments 42
2.4 The Structure of Agents 47
Summary 60
Bibliographical and Historical Notes 60
II Problem-solving
3 Solving Problems by Searching 63
3.1 Problem-Solving Agents 63
3.2 Example Problems 66
3.3 Search Algorithms 71
3.4 Uninformed Search Strategies 76
3.5 Informed (Heuristic) Search Strategies 84
3.6 Heuristic Functions 97
Summary 104
Bibliographical and Historical Notes 106
4 Search in Complex Environments 110
4.1 Local Search and Optimization Problems 110
4.2 Local Search in Continuous Spaces 119
4.3 Search with Nondeterministic Actions 122
4.4 Search in Partially Observable Environments 126
4.5 Online Search Agents and Unknown Environments 134
Summary 141
Bibliographical and Historical Notes 142
5 Adversarial Search and Games 146
5.1 Game Theory 146
5.2 Optimal Decisions in Games 148
5.3 Heuristic Alpha–Beta Tree Search 156
5.4 Monte Carlo Tree Search 161
5.5 Stochastic Games 164
5.6 Partially Observable Games 168
5.7 Limitations of Game Search Algorithms 173
Summary 174
Bibliographical and Historical Notes 175
6 Constraint Satisfaction Problems 180
6.1 Defining Constraint Satisfaction Problems 180
6.2 Constraint Propagation: Inference in CSPs 185
6.3 Backtracking Search for CSPs 1916.4 Local Search for CSPs 197
6.5 The Structure of Problems 199
Summary 203
Bibliographical and Historical Notes 204
III Knowledge, reasoning, and planning
7 Logical Agents 208
7.1 Knowledge-Based Agents 209
7.2 The Wumpus World 210
7.3 Logic 214
7.4 Propositional Logic: A Very Simple Logic 217
7.5 Propositional Theorem Proving 222
7.6 Effective Propositional Model Checking 232
7.7 Agents Based on Propositional Logic 237
Summary 246
Bibliographical and Historical Notes 247
8 First-Order Logic 251
8.1 Representation Revisited 251
8.2 Syntax and Semantics of First-Order Logic 256
8.3 Using First-Order Logic 265
8.4 Knowledge Engineering in First-Order Logic 271
Summary 277
Bibliographical and Historical Notes 278
9 Inference in First-Order Logic 280
9.1 Propositional vs. First-Order Inference 280
9.2 Unification and First-Order Inference 2829.3 Forward Chaining 286
9.4 Backward Chaining 293
9.5 Resolution 298
Summary 309
Bibliographical and Historical Notes 310
10 Knowledge Representation 314
10.1 Ontological Engineering 314
10.2 Categories and Objects 317
10.3 Events 322
10.4 Mental Objects and Modal Logic 326
10.5 Reasoning Systems for Categories 329
10.6 Reasoning with Default Information 333
Summary 337
Bibliographical and Historical Notes 338
11 Automated Planning 344
11.1 Definition of Classical Planning 344
11.2 Algorithms for Classical Planning 348
11.3 Heuristics for Planning 353
11.4 Hierarchical Planning 356
11.5 Planning and Acting in Nondeterministic Domains 365
11.6 Time, Schedules, and Resources 374
11.7 Analysis of Planning Approaches 378
Summary 379
Bibliographical and Historical Notes 380

IV Uncertain knowledge and reasoning
12 Quantifying Uncertainty 385
12.1 Acting under Uncertainty 385
12.2 Basic Probability Notation 388
12.3 Inference Using Full Joint Distributions 395
12.4 Independence 397
12.5 Bayes’ Rule and Its Use 399
12.6 Naive Bayes Models 402
12.7 The Wumpus World Revisited 404
Summary 407
Bibliographical and Historical Notes 408
13 Probabilistic Reasoning 412
13.1 Representing Knowledge in an Uncertain Domain 412
13.2 The Semantics of Bayesian Networks 414
13.3 Exact Inference in Bayesian Networks 427
13.4 Approximate Inference for Bayesian Networks 435
13.5 Causal Networks 449
Summary 453
Bibliographical and Historical Notes 454
14 Probabilistic Reasoning over Time 461
14.1 Time and Uncertainty 461
14.2 Inference in Temporal Models 465
14.3 Hidden Markov Models 473
14.4 Kalman Filters 479
14.5 Dynamic Bayesian Networks 485Summary 496
Bibliographical and Historical Notes 497
15 Probabilistic Programming 500
15.1 Relational Probability Models 501
15.2 Open-Universe Probability Models 507
15.3 Keeping Track of a Complex World 514
15.4 Programs as Probability Models 519
Summary 523
Bibliographical and Historical Notes 524
16 Making Simple Decisions 528
16.1 Combining Beliefs and Desires under Uncertainty 528
16.2 The Basis of Utility Theory 529
16.3 Utility Functions 532
16.4 Multiattribute Utility Functions 540
16.5 Decision Networks 544
16.6 The Value of Information 547
16.7 Unknown Preferences 553
Summary 557
Bibliographical and Historical Notes 557
17 Making Complex Decisions 562
17.1 Sequential Decision Problems 562
17.2 Algorithms for MDPs 572
17.3 Bandit Problems 581
17.4 Partially Observable MDPs 588
17.5 Algorithms for Solving POMDPs 590Summary 595
Bibliographical and Historical Notes 596
18 Multiagent Decision Making 599
18.1 Properties of Multiagent Environments 599
18.2 Non-Cooperative Game Theory 605
18.3 Cooperative Game Theory 626
18.4 Making Collective Decisions 632
Summary 645
Bibliographical and Historical Notes 646
V Machine Learning
19 Learning from Examples 651
19.1 Forms of Learning 651
19.2 Supervised Learning 653
19.3 Learning Decision Trees 657
19.4 Model Selection and Optimization 665
19.5 The Theory of Learning 672
19.6 Linear Regression and Classification 676
19.7 Nonparametric Models 686
19.8 Ensemble Learning 696
19.9 Developing Machine Learning Systems 704
Summary 714
Bibliographical and Historical Notes 715
20 Learning Probabilistic Models 721
20.1 Statistical Learning 721
20.2 Learning with Complete Data 72420.3 Learning with Hidden Variables: The EM Algorithm 737
Summary 746
Bibliographical and Historical Notes 747
21 Deep Learning 750
21.1 Simple Feedforward Networks 751
21.2 Computation Graphs for Deep Learning 756
21.3 Convolutional Networks 760
21.4 Learning Algorithms 765
21.5 Generalization 768
21.6 Recurrent Neural Networks 772
21.7 Unsupervised Learning and Transfer Learning 775
21.8 Applications 782
Summary 784
Bibliographical and Historical Notes 785
22 Reinforcement Learning 789
22.1 Learning from Rewards 789
22.2 Passive Reinforcement Learning 791
22.3 Active Reinforcement Learning 797
22.4 Generalization in Reinforcement Learning 803
22.5 Policy Search 810
22.6 Apprenticeship and Inverse Reinforcement Learning 812
22.7 Applications of Reinforcement Learning 815
Summary 818
Bibliographical and Historical Notes 819

VI Communicating, perceiving, and acting
23 Natural Language Processing 823
23.1 Language Models 823
23.2 Grammar 833
23.3 Parsing 835
23.4 Augmented Grammars 841
23.5 Complications of Real Natural Language 845
23.6 Natural Language Tasks 849
Summary 850
Bibliographical and Historical Notes 851
24 Deep Learning for Natural Language Processing 856
24.1 Word Embeddings 856
24.2 Recurrent Neural Networks for NLP 860
24.3 Sequence-to-Sequence Models 864
24.4 The Transformer Architecture 868
24.5 Pretraining and Transfer Learning 871
24.6 State of the art 875
Summary 878
Bibliographical and Historical Notes 878
25 Computer Vision 881
25.1 Introduction 881
25.2 Image Formation 882
25.3 Simple Image Features 888
25.4 Classifying Images 895
25.5 Detecting Objects 89925.6 The 3D World 901
25.7 Using Computer Vision 906
Summary 919
Bibliographical and Historical Notes 920
26 Robotics 925
26.1 Robots 925
26.2 Robot Hardware 926
26.3 What kind of problem is robotics solving? 930
26.4 Robotic Perception 931
26.5 Planning and Control 938
26.6 Planning Uncertain Movements 956
26.7 Reinforcement Learning in Robotics 958
26.8 Humans and Robots 961
26.9 Alternative Robotic Frameworks 968
26.10 Application Domains 971
Summary 974
Bibliographical and Historical Notes 975
VII Conclusions
27 Philosophy, Ethics, and Safety of AI 981
27.1 The Limits of AI 981
27.2 Can Machines Really Think? 984
27.3 The Ethics of AI 986
Summary 1005
Bibliographical and Historical Notes 1006
28 The Future of AI 101228.1 AI Components 1012
28.2 AI Architectures 1018
A Mathematical Background 1023
A.1 Complexity Analysis and O() Notation 1023
A.2 Vectors, Matrices, and Linear Algebra 1025
A.3 Probability Distributions 1027
Bibliographical and Historical Notes 1029
B Notes on Languages and Algorithms 1030
B.1 Defining Languages with Backus–Naur Form (BNF) 1030
B.2 Describing Algorithms with Pseudocode 1031
B.3 Online Supplemental Material 1032
Bibliography 1033
Index 1069