Muutke küpsiste eelistusi

Artificial Intelligence: With an Introduction to Machine Learning, Second Edition 2nd edition [Kõva köide]

(University of Pittsburgh, Pennsylvania, USA), (Northeastern Illinois University, Illinois, USA)
Teised raamatud teemal:
  • Kõva köide
  • Hind: 133,88 €*
  • * hind on lõplik, st. muud allahindlused enam ei rakendu
  • Tavahind: 178,50 €
  • Säästad 25%
  • Raamatu kohalejõudmiseks kirjastusest kulub orienteeruvalt 3-4 nädalat
  • Kogus:
  • Lisa ostukorvi
  • Tasuta tarne
  • Tellimisaeg 2-4 nädalat
  • Lisa soovinimekirja
Artificial Intelligence: With an Introduction to Machine Learning, Second Edition 2nd editionSuurem pilt
Teised raamatud teemal:
Teised raamatud sellest sooduspakkumisest: Kevadine sooduspakkumine - kuni 31. maini üle miljoni raamatu kuni 25% soodsamalt
Püsilink: https://www.kriso.ee/db/9781138502383.html
Märksõnad:
The first edition of this popular textbook, Contemporary Artificial Intelligence, provided an accessible and student friendly introduction to AI. This fully revised and expanded update, Artificial Intelligence: With an Introduction to Machine Learning, Second Edition, retains the same accessibility and problem-solving approach, while providing new material and methods.

The book is divided into five sections that focus on the most useful techniques that have emerged from AI. The first section of the book covers logic-based methods, while the second section focuses on probability-based methods. Emergent intelligence is featured in the third section and explores evolutionary computation and methods based on swarm intelligence. The newest section comes next and provides a detailed overview of neural networks and deep learning. The final section of the book focuses on natural language understanding.

Suitable for undergraduate and beginning graduate students, this class-tested textbook provides students and other readers with key AI methods and algorithms for solving challenging problems involving systems that behave intelligently in specialized domains such as medical and software diagnostics, financial decision making, speech and text recognition, genetic analysis, and more.

Arvustused

At many universities courses on arti cial intelligence (AI) are offered, mainly for computer science students. This is very often a bit optimistic since this field also requires a sound mathematical background. Furthermore, there is now an increasing rumor about the problems, dangers etc. that may appear. In this field this textbook is an excellent contribution to avoid these discussions and make artificial intelligence more and more a practicable field!

-Christian Postho, St. Augustine

Preface xi
About the Authors xiii
1 Introduction to Artificial Intelligence
1(8)
1.1 History of Artificial Intelligence
2(5)
1.1.1 What Is Artificial Intelligence?
2(2)
1.1.2 Emergence of AI
4(1)
1.1.3 Cognitive Science and AI
4(1)
1.1.4 Logical Approach to AI
4(1)
1.1.5 Knowledge-Based Systems
5(1)
1.1.6 Probabilistic Approach to AI
6(1)
1.1.7 Evolutionary Computation and Swarm Intelligence
6(1)
1.1.8 Neural Networks & Deep Learning
7(1)
1.1.9 A Return to Creating HAL
7(1)
1.2 Outline of This Book
7(2)
I Logical Intelligence 9(104)
2 Propositional Logic
11(42)
2.1 Basics of Propositional Logic
12(12)
2.1.1 Syntax
12(1)
2.1.2 Semantics
13(4)
2.1.3 Tautologies and Logical Implication
17(1)
2.1.4 Logical Arguments
18(3)
2.1.5 Derivation Systems
21(3)
2.2 Resolution
24(6)
2.2.1 Normal Forms
25(1)
2.2.2 Derivations Using Resolution
26(4)
2.2.3 Resolution Algorithm
30(1)
2.3 Artificial Intelligence Applications
30(18)
2.3.1 Knowledge-Based Systems
30(11)
2.3.2 Wumpus World
41(7)
2.4 Discussion and Further Reading
48(5)
3 First-Order Logic
53(24)
3.1 Basics of First-Order Logic
53(15)
3.1.1 Syntax
54(2)
3.1.2 Semantics
56(4)
3.1.3 Validity and Logical Implication
60(2)
3.1.4 Derivation Systems
62(3)
3.1.5 Modus Ponens for First-Order Logic
65(3)
3.2 Artificial Intelligence Applications
68(5)
3.2.1 Wumpus World Revisited
69(1)
3.2.2 Planning
69(4)
3.3 Discussion and Further Reading
73(4)
4 Certain Knowledge Representation
77(12)
4.1 Taxonomic Knowledge
78(2)
4.1.1 Semantic Nets
78(1)
4.1.2 Model of Human Organization of Knowledge
79(1)
4.2 Frames
80(4)
4.2.1 Frame Data Structure
80(1)
4.2.2 Planning a Trip Using Frames
81(3)
4.3 Nonmonotonic Logic
84(2)
4.3.1 Circumscription
84(1)
4.3.2 Default Logic
85(1)
4.3.3 Difficulties
86(1)
4.4 Discussion and Further Reading
86(3)
5 Learning Deterministic Models
89(24)
5.1 Supervised Learning
89(1)
5.2 Regression
90(6)
5.2.1 Simple Linear Regression
91(2)
5.2.2 Multiple Linear Regression
93(1)
5.2.3 Overfitting and Cross Validation
94(2)
5.3 Parameter Estimation
96(6)
5.3.1 Estimating the Parameters for Simple Linear Regression
96(2)
5.3.2 Gradient Descent
98(2)
5.3.3 Logistic Regression and Gradient Descent
100(1)
5.3.4 Stochastic Gradient Descent
101(1)
5.4 Learning a Decision Tree
102(13)
5.4.1 Information Theory
102(4)
5.4.2 Information Gain and the ID3 Algorithm
106(2)
5.4.3 Overfitting
108(5)
II Probabilistic Intelligence 113(236)
6 Probability
115(30)
6.1 Probability Basics
117(6)
6.1.1 Probability Spaces
117(3)
6.1.2 Conditional Probability and Independence
120(2)
6.1.3 Bayes' Theorem
122(1)
6.2 Random Variables
123(8)
6.2.1 Probability Distributions of Random Variables
123(5)
6.2.2 Independence of Random Variables
128(3)
6.3 Meaning of Probability
131(4)
6.3.1 Relative Frequency Approach to Probability
132(2)
6.3.2 Subjective Approach to Probability
134(1)
6.4 Random Variables in Applications
135(4)
6.5 Probability in the Wumpus World
139(6)
7 Uncertain Knowledge Representation
145(36)
7.1 Intuitive Introduction to Bayesian Networks
147(2)
7.2 Properties of Bayesian Networks
149(5)
7.2.1 Definition of a Bayesian Network
149(3)
7.2.2 Representation of a Bayesian Network
152(2)
7.3 Causal Networks as Bayesian Networks
154(6)
7.3.1 Causality
154(1)
7.3.2 Causality and the Markov Condition
155(4)
7.3.3 Markov Condition without Causality
159(1)
7.4 Inference in Bayesian Networks
160(5)
7.4.1 Examples of Inference
160(2)
7.4.2 Inference Algorithms and Packages
162(1)
7.4.3 Inference Using Netica
163(2)
7.5 Networks with Continuous Variables
165(5)
7.5.1 Gaussian Bayesian Networks
165(3)
7.5.2 Hybrid Networks
168(2)
7.6 Obtaining the Probabilities
170(4)
7.6.1 Difficulty Inherent in Multiple Parents
170(1)
7.6.2 Basic Noisy OR-Gate Model
170(2)
7.6.3 Leaky Noisy OR-Gate Model
172(2)
7.6.4 Further Models
174(1)
7.7 Large-Scale Application: Promedas
174(7)
8 Advanced Properties of Bayesian Networks
181(20)
8.1 Entailed Conditional Independencies
182(6)
8.1.1 Examples of Entailed Conditional Independencies
182(3)
8.1.2 d-Separation
185(3)
8.2 Faithfulness
188(3)
8.2.1 Unfaithful Probability Distributions
188(2)
8.2.2 Faithfulness Condition
190(1)
8.3 Markov Equivalence
191(1)
8.4 Markov Blankets and Boundaries
192(9)
9 Decision Analysis
201(56)
9.1 Decision Trees
202(14)
9.1.1 Simple Examples
202(3)
9.1.2 Solving More Complex Decision Trees
205(11)
9.2 Influence Diagrams
216(15)
9.2.1 Representing Decision Problems with Influence Diagrams
216(6)
9.2.2 Solving Influence Diagrams
222(1)
9.2.3 Techniques for Solving Influence Diagrams
222(4)
9.2.4 Solving Influence Diagrams Using Netica
226(5)
9.3 Modeling Risk Preferences
231(2)
9.3.1 Exponential Utility Function
231(1)
9.3.2 Assessing r
232(1)
9.4 Analyzing Risk Directly
233(6)
9.4.1 Using the Variance to Measure Risk
233(2)
9.4.2 Risk Profiles
235(1)
9.4.3 Dominance
236(3)
9.5 Good Decision versus Good Outcome
239(1)
9.6 Sensitivity Analysis
239(2)
9.7 Value of Information
241(4)
9.7.1 Expected Value of Perfect Information
242(2)
9.7.2 Expected Value of Imperfect Information
244(1)
9.8 Discussion and Further Reading
245(12)
9.8.1 Academics
246(1)
9.8.2 Business and Finance
247(1)
9.8.3 Capital Equipment
247(1)
9.8.4 Computer Games
247(1)
9.8.5 Computer Vision
247(1)
9.8.6 Computer Software
247(1)
9.8.7 Medicine
248(1)
9.8.8 Natural Language Processing
248(1)
9.8.9 Planning
248(1)
9.8.10 Psychology
248(1)
9.8.11 Reliability Analysis
248(1)
9.8.12 Scheduling
249(1)
9.8.13 Speech Recognition
249(1)
9.8.14 Vehicle Control and Malfunction Diagnosis
249(8)
10 Learning Probabilistic Model Parameters
257(18)
10.1 Learning a Single Parameter
257(4)
10.1.1 Binomial Random Variables
258(2)
10.1.2 Multinomial Random Variables
260(1)
10.2 Learning Parameters in a Bayesian Network
261(5)
10.2.1 Procedure for Learning Parameters
262(1)
10.2.2 Equivalent Sample Size
263(3)
10.3 Learning Parameters with Missing Data
266(9)
11 Learning Probabilistic Model Structure
275(56)
11.1 Structure Learning Problem
276(1)
11.2 Score-Based Structure Learning
276(27)
11.2.1 Bayesian Score
276(7)
11.2.2 BIC Score
283(1)
11.2.3 Consistent Scoring Criteria
284(1)
11.2.4 How Many DAGs Must We Score?
285(1)
11.2.5 Using the Learned Network to Do Inference
285(1)
11.2.6 Learning Structure with Missing Data
286(7)
11.2.7 Approximate Structure Learning
293(4)
11.2.8 Model Averaging
297(3)
11.2.9 Approximate Model Averaging
300(3)
11.3 Constraint-Based Structure Learning
303(5)
11.3.1 Learning a DAG Faithful to P
303(4)
11.3.2 Learning a DAG in which P Is Embedded Faithfully
307(1)
11.4 Application: MENTOR
308(3)
11.4.1 Developing the Network
308(2)
11.4.2 Validating MENTOR
310(1)
11.5 Software Packages for Learning
311(1)
11.6 Causal Learning
312(8)
11.6.1 Causal Faithfulness Assumption
312(2)
11.6.2 Causal Embedded Faithfulness Assumption
314(3)
11.6.3 Application: College Student Retention Rate
317(3)
11.7 Class Probability Trees
320(5)
11.7.1 Theory of Class Probability Trees
320(2)
11.7.2 Application to Targeted Advertising
322(3)
11.8 Discussion and Further Reading
325(6)
11.8.1 Biology
325(1)
11.8.2 Business and Finance
326(1)
11.8.3 Causal Learning
326(1)
11.8.4 Data Mining
326(1)
11.8.5 Medicine
326(1)
11.8.6 Weather Forecasting
326(5)
12 Unsupervised Learning and Reinforcement Learning
331(18)
12.1 Unsupervised Learning
331(2)
12.1.1 Clustering
331(2)
12.1.2 Automated Discovery
333(1)
12.2 Reinforcement Learning
333(12)
12.2.1 Multi-Armed Bandit Algorithms
333(3)
12.2.2 Dynamic Networks
336(9)
12.3 Discussion and Further Reading
345(4)
III Emergent Intelligence 349(38)
13 Evolutionary Computation
351(26)
13.1 Genetics Review
352(2)
13.2 Genetic Algorithms
354(10)
13.2.1 Algorithm
354(1)
13.2.2 Illustrative Example
355(2)
13.2.3 Traveling Salesperson Problem
357(7)
13.3 Genetic Programming
364(9)
13.3.1 Illustrative Example
365(2)
13.3.2 Artificial Ant
367(3)
13.3.3 Application to Financial Trading
370(3)
13.4 Discussion and Further Reading
373(4)
14 Swarm Intelligence
377(10)
14.1 Ant System
377(4)
14.1.1 Real Ant Colonies
378(1)
14.1.2 Artificial Ants for Solving the TSP
378(3)
14.2 Flocks
381(2)
14.3 Discussion and Further Reading
383(4)
IV Neural Intelligence 387(26)
15 Neural Networks and Deep Learning
389(24)
15.1 The Perceptron
389(6)
15.1.1 Learning the Weights for a Perceptron
391(3)
15.1.2 The Perceptron and Logistic Regression
394(1)
15.2 Feedforward Neural Networks
395(8)
15.2.1 Modeling XOR
395(3)
15.2.2 Example with Two Hidden Layers
398(3)
15.2.3 Structure of a Feedforward Neural Network
401(2)
15.3 Activation Functions
403(4)
15.3.1 Output Nodes
403(2)
15.3.2 Hidden Nodes
405(2)
15.4 Application to Image Recognition
407(1)
15.5 Discussion and Further Reading
407(6)
V Language Understanding 413(24)
16 Natural Language Understanding
415(22)
16.1 Parsing
417(13)
16.1.1 Recursive Parser
418(2)
16.1.2 Ambiguity
420(2)
16.1.3 Dynamic Programming Parser
422(4)
16.1.4 Probabilistic Parser
426(2)
16.1.5 Obtaining Probabilities for a PCFG
428(1)
16.1.6 Lexicalized PCFG
428(2)
16.2 Semantic Interpretation
430(1)
16.3 Concept/Knowledge Interpretation
431(1)
16.4 Information Extraction
432(3)
16.4.1 Applications of Information Extraction
432(1)
16.4.2 Architecture for an Information Extraction System
433(2)
16.5 Discussion and Further Reading
435(2)
References 437(22)
Index 459
Richard E. Neapolitan is professor emeritus of computer science at Northeastern Illinois University and a former professor of bioinformatics at Northwestern University. He is currently president of Bayesian Network Solutions. His research interests include probability and statistics, decision support systems, cognitive science, and applications of probabilistic modeling to fields such as medicine, biology, and finance. Dr. Neapolitan is a prolific author and has published in the most prestigious journals in the broad area of reasoning under uncertainty. He has previously written five books, including the seminal 1989 Bayesian network text Probabilistic Reasoning in Expert Systems; Learning Bayesian Networks (2004); Foundations of Algorithms (1996, 1998, 2003, 2010, 2015), which has been translated into three languages; Probabilistic Methods for Financial and Marketing Informatics (2007); and Probabilistic Methods for Bioinformatics (2009). His approach to textbook writing is distinct in that he introduces a concept or methodology with simple examples, and then provides the theoretical underpinning. As a result, his books have the reputation for making difficult material easy to understand without sacrificing scientific rigor.

Xia Jiang is an associate professor in the Department of Biomedical Informatics at the University of Pittsburgh School of Medicine. She has over 16 years of teaching and research experience using artificial intelligence, machine learning, Bayesian networks, and causal learning to model and solve problems in biology, medicine, and translational science. Dr. Jiang pioneered the application of Bayesian networks and information theory to the task of learning causal interactions such as genetic epistasis from data, and she has conducted innovative research in the areas of cancer informatics, probabilistic medical decision support, and biosurveillance. She is the coauthor of the book Probabilistic Methods for Financial and Marketing Informatics (2007).