Solutions for Artificial Intelligence A Modern Approach, 4th Edition by Russell (All Chapters included)

SOLUTIONS & INSTRUCTOR MANUAL
Artificial Intelligence: A Modern Approach, 4th Edition
by Peter Norvig and Stuart Russell, Chapters 1 – 28

,Artificial Intelligence QWE




1 Introduction ...
QWEQWEQWEQWE QWE QWE QWE




2 Intelligent Agents ...
QWEQWEQWEQWE QWE QWE QWE QWE




II Problem-solving
QWE




3 Solving Problems by Searching ...
QWEQWEQWEQWE QWE QWE QWE QWE QWE QWE




4 Search in Complex Environments ...
QWEQWEQWEQWE QWE QWE QW E QWE QWE QWE




5 Adversarial Search and Games ...
QWEQWEQWEQWE QWE QWE QWE QWE QW E QWE




6 Constraint Satisfaction Problems ...
QWEQWEQWEQWE QWE QWE QWE QWE QWE




III Knowledge, reasoning, and planning
QWE QWE QWE QW E




7 Logical Agents ...
QWEQWEQWEQWE QWE QWE QWE QWE




8 First-Order Logic ...
QWEQWEQWEQWE QWE QWE QWE QWE




9 Inference in First-Order Logic ...
QWEQWEQWEQWE QWE QW E QWE QWE QWE




10 Knowledge Representation ...
QWEQWEQWEQWE QWE QWE QWE QWE




11 Automated Planning ...
QWEQWEQWEQWE QWE QWE QW E QWE




IV Uncertain knowledge and reasoning
QWE QWE QWE QWE




12 Quantifying Uncertainty ...
QWEQWEQWEQWE QWE QWE QWE QWE




13 Probabilistic Reasoning ...
QWEQWEQWEQWE QWE QWE QWE QWE




14 Probabilistic Reasoning over Time ...
QWEQWEQWEQWE QWE QWE QWE QWE QWE QWE




15 Probabilistic Programming ...
QWEQWEQWEQWE QWE QWE QWE QWE




16 Making Simple Decisions ...
QWEQWEQWEQWE QWE QWE QWE QWE QWE




17 Making Complex Decisions ...
QWEQWEQWEQWE QWE QWE QWE QWE QWE




18 Multiagent Decision Making ...
QWEQWEQWEQWE QWE QW E QWE QWE QWE




V Machine Learning
QWE QWE

, 19 Learning from Examples ...
QWEQWEQWEQWE QWE QWE QWE QWE QWE




20 Learning Probabilistic Models ...
QWEQWEQWEQWE QWE QWE QWE QWE QWE




21 Deep Learning ...
QWEQWEQWEQWE QWE QW E QWE QWE




22 Reinforcement Learning ...
QWEQWEQWEQWE QWE QWE QWE QW E




VI Communicating, perceiving, and acting
QWE QWE QWE QWE




23 Natural Language Processing ...
QWEQWEQWEQWE QWE QWE QW E QWE QWE




24 Deep Learning for Natural Language Processing ...
QWEQWEQWEQWE QWE QW E QWE QW E QWE QW E QWE QWE




25 Computer Vision ...
QWEQWEQWEQWE QWE QWE QW E QW E




26 Robotics ...
QWEQWEQWEQWE QWE QWE QW E




VII Conclusions QWE




27 Philosophy, Ethics, and Safety of AI ...
QWEQWEQWEQWE QWE QWE QWE QW E QWE QWE QWE QWE




28 The Future of AI
QWEQWEQWEQWE QWE QWE QWE QWE

, EXERCISES Q W E Q W E




1
INTRODUCTION
Notethatformanyofthequestionsinthischapter,wegivereferenceswhereanswerscanbe found r
QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE




ather than writing them out—the full answers would be far too long.
QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE




1.1 What Is AI? Q W E QWE QWE




Exercise1.1.#DEFA W
Q
E




Define in your own words: (a) intelligence, (b) artificial intelligence, (c) agent, (d) ra- QWE QWE QWE QW E Q W E QWE QWE QWE QWE QWE QWE QWE QWE




tionality, (e) logical reasoning.
QWE QWE QWE QWE




a. Dictionary definitions of intelligence talk about “the capacity to acquire and apply knowl QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE




edge” or “the faculty of thought and reason” or “the ability to comprehend and profit from QWE QWE QWE QWE QW E QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QW E




experience.” These are all reasonable answers, but if we want something quantifiable w Q W E QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE




e would use something like “the ability to act successfullyacross a wide range of objectives i
QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE




n complex environments.” QWE QWE




b. We define artificial intelligence as the study and construction of agent programs that per
QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE




form well in a given class of environments, for a given agent architecture; they do the righ
QWE QWE QWE QWE QW E QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE




t thing. An important part of that is dealing with the uncertainty of what the current state i
QWE Q W E QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QW E QWE QWE QWE QWE




s, what the outcome of possible actions might be, and what is it that we really desire.
QWE QWE QWE QWE QW E QWE QWE QW E QWE QWE QWE QWE QWE QWE QWE QWE




c. We define an agent as an entity that takes action in response to percepts from an envi-
QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE QWE




ronment. QWE




d. We define rationality as the property of a system which does the “right thing” given what
QWE QWE QWE QWE QWE QWE QWE QW E QWE QWE QWE QWE QWE QWE QW E QW




it knows. See Section 2.2 for a more complete discussion. The basic concept is perfectr
E QWE Q W E QWE QWE QWE QWE QWE QW E QWE Q W E QWE QWE QWE QWE QWE




ationality;Section??describestheimpossibilityofachievingperfectrational- QWE QWE WE
Q QW
E QW
E QW
E QW
E QW
E QW
E




ity and proposes an alternative definition.
QWE QW E QWE QWE QWE QW E




e. Wedefinelogical reasoning astheaprocessofderivingnewsentencesfromold,such thatth
QWE QWE QWE QWE QW
E QW
E QWE QWE QW
E QWE QWE QW
E QWE QWE QWE E
QW




enewsentencesarenecessarilytrueiftheoldonesaretrue. (Noticethatdoesnot refertoanysp
E
QW E
QW E
QW E
QW E
QW E
QW E
QW E
QW E
QW E
QW E
QW QWE E
QW E
QW E
QW QWE E
QW E
QW E
QW




ecificsyntaxorformallanguage,butitdoesrequireawell-definednotion of truth.) E
QW E
QW E
QW E
QW EQW E
QW E
QW E
QW E
QW E
QW E
QW QWE QWE




Exercise1.1.#TURI W
Q
E




Read Turing’s original paper on AI (Turing, 1950). In the paper, he discusses several objectio
QWE QWE QWE QWE QWE QWE QWE Q W E QWE QWE QWE QWE QWE QWE




nstohisproposedenterpriseandhistestforintelligence. Whichobjectionsstillcarry
E
QW E
QW E
QW E
QW E
QW E
QW E
QW E
QW E
QW QW E E
QW E
QW E
QW




© 2023 Pearson Education, Hoboken, NJ. All rights reserved.
QWE QWE QWE QWE QWE QWE QWE QWE