Solution Manual for Artificial Intelligence A Modern Approach, 4th Edition by Stuart Russell and Peter Norvig.

SOLUTIONS & INSTRUCTOR MANUAL
cn cn cn




Artificial Intelligence: A Modern Approach, 4th Edition
cn cn cn cn cn cn




by Peter Norvig and Stuart Russell, Chapters 1 – 28
cn cn cn cn cn cn cn cn cn cn

,Artificial Intelligence cn




cn cn cn cn 1 Introduction ...
cn cn cn




cn cn cn cn 2 Intelligent Agents ...
cn cn cn cn




II Problem-solving cn




cn cn cn cn 3 Solving Problems by Searching ...
cn cn cn cn cn cn




cn cn cn cn 4 Search in Complex Environments ...
cn cn cn cn cn cn




cn cn cn cn 5 Adversarial Search and Games ...
cn cn cn cn cn cn




cn cn cn cn 6 Constraint Satisfaction Problems ...
cn cn cn cn cn




III Knowledge, reasoning, and planning
cn cn cn cn




cn cn cn cn 7 Logical Agents ...
cn cn cn cn




cn cn cn cn 8 First-Order Logic ...
cn cn cn cn




cn cn cn cn 9 Inference in First-Order Logic ...
cn cn cn cn cn




cn cn cn cn 10 Knowledge Representation ...
cn cn cn cn




cn cn cn cn 11 Automated Planning ...cn cn cn cn




IV Uncertain knowledge and reasoning
cn cn cn cn




cn cn cn cn 12 Quantifying Uncertainty ...
cn cn cn cn




cn cn cn cn 13 Probabilistic Reasoning ...
cn cn cn cn




cn cn cn cn 14 Probabilistic Reasoning over Time ...
cn cn cn cn cn cn




cn cn cn cn 15 Probabilistic Programming ...
cn cn cn cn




cn cn cn cn 16 Making Simple Decisions ...
cn cn cn cn cn




cn cn cn cn 17 Making Complex Decisions ...
cn cn cn cn cn




cn cn cn cn 18 Multiagent Decision Making ...
cn cn cn cn cn




V Machine Learning
cn cn

,cn cn cn cn 19 Learning from Examples ...
cn cn cn cn cn




cn cn cn cn 20 Learning Probabilistic Models ...
cn cn cn cn cn




cn cn cn cn 21 Deep Learning ...
cn cn cn cn




cn cn cn cn 22 Reinforcement Learning ...
cn cn cn cn




VI Communicating, perceiving, and acting
cn cn cn cn




cn cn cn cn 23 Natural Language Processing ...
cn cn cn cn cn




cn cn cn cn 24 Deep Learning for Natural Language Processing ...
cn cn cn cn cn cn cn cn




cn cn cn cn 25 Computer Vision ...
cn cn cn cn




cn cn cn cn 26 Robotics ...
cn cn cn




VII Conclusions cn




cn cn cn cn 27 Philosophy, Ethics, and Safety of AI ...
cn cn cn cn cn cn cn cn




cn cn cn cn 28 The Future of AI
cn cn cn cn

, EXERCISES c n c n




1
INTRODUCTION
Notethatformanyofthequestionsinthischapter, wegivereferenceswhereanswerscanbe found
cn cn cn cn cn cn cn cn cn cn cn cn cn cn cn cn cn




rather than writing them out—the full answers would be far too long.
cn cn cn cn cn cn cn cn cn cn cn cn




1.1 What Is AI?c n cn cn




Exercise1.1.#DEFA cn




Define in your own words: (a) intelligence, (b) artificial intelligence, (c) agent, (d) ra- cn cn cn cn c n cn cn cn cn cn cn cn cn




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




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




knowledge” or “the faculty of thought and reason” or “the ability to comprehend and
cn cn cn cn cn cn cn cn cn cn cn cn cn cn




profit from experience.” These are all reasonable answers, but if we want something
cn cn cn c n cn cn cn cn cn cn cn cn cn




quantifiable we would use something like “the ability to act successfully across a wide range
cn cn cn cn cn cn cn cn cn cn cn cn cn cn cn




of objectives in complex environments.”
cn cn cn cn cn




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




perform well in a given class of environments, for a given agent architecture; they do the
cn cn cn cn cn cn cn cn cn cn cn cn cn cn cn cn




right thing. An important part of that is dealing with the uncertainty of what the current
cn cn c n cn cn cn cn cn cn cn cn cn cn cn cn cn




state is, what the outcome of possible actions might be, and what is it that we really desire.
cn cn cn cn cn cn cn cn cn cn cn cn cn cn cn cn cn cn




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




ronment. cn




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




it knows. See Section 2.2 for a more complete discussion. The basic concept is perfect
cn cn c n cn cn cn cn cn cn cn c n cn cn cn cn




rationality; Section ?? describes the impossibility of achieving perfect rational- ity and
cn cn cn cn cn cn cn cn cn cn cn cn




proposes an alternative definition.
cn cn cn cn




e. Wedefine logicalreasoning astheaprocess ofderiving newsentences from old,such thatthe
cn cn cn cn cn cn cn cn cn cn cn cn cn cn cn cn




new sentences are necessarily true if the old ones are true. (Notice that does not refer to any
cn cn cn cn cn cn cn cn cn cn cn cn cn cn cn cn cn cn




specificsyntaxorformallanguage,butitdoesrequireawell-definednotion of truth.)
cn cn cn cn cn cn cn cn cn cn cn cn cn cn




Exercise1.1.#TURI cn




Read Turing’s original paper on AI (Turing, 1950). In the paper, he discusses several
cn cn cn cn cn cn cn c n cn cn cn cn cn




objectionstohisproposedenterpriseandhistestforintelligence. Whichobjectionsstillcarry
cn cn cn cn cn cn cn cn cn cn cn cn cn cn




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