n n n
Artificial Intelligence: A Modern Approach, 4th Edition
n n n n n n
by Peter Norvig and Stuart Russell, Chapters 1 – 28
n n n n n n n n n n
,Artificial Intelligence
n
n n n n 1 Introduction ...
n n n
n n n n 2 Intelligent Agents ...
n n n n
II Problem-solving
n
n n n n 3 Solving Problems by Searching ...
n n n n n n
n n n n 4 Search in Complex Environments ...
n n n n n n
n n n n 5 Adversarial Search and Games ...
n n n n n n
n n n n 6 Constraint Satisfaction Problems ...
n n n n n
III Knowledge, reasoning, and planning
n n n n
n n n n 7 Logical Agents ...
n n n n
n n n n 8 First-Order Logic ...
n n n n
n n n n 9 Inference in First-Order Logic ...
n n n n n
n n n n 10 Knowledge Representation ...
n n n n
n n n n 11 Automated Planning ...
n n n n
IV Uncertain knowledge and reasoning
n n n n
n n n n 12 Quantifying Uncertainty ...
n n n n
n n n n 13 Probabilistic Reasoning ...
n n n n
n n n n 14 Probabilistic Reasoning over Time ...
n n n n n n
n n n n 15 Probabilistic Programming ...
n n n n
n n n n 16 Making Simple Decisions ...
n n n n n
n n n n 17 Making Complex Decisions ...
n n n n n
n n n n 18 Multiagent Decision Making ...
n n n n n
,V Machine Learning
n n
n n n n 19 Learning from Examples ...
n n n n n
n n n n 20 Learning Probabilistic Models ...
n n n n n
n n n n 21 Deep Learning ...
n n n n
n n n n 22 Reinforcement Learning ...
n n n n
VI Communicating, perceiving, and acting
n n n n
n n n n 23 Natural Language Processing ...
n n n n n
n n n n 24 Deep Learning for Natural Language Processing ...
n n n n n n n n
n n n n 25 Computer Vision ...
n n n n
n n n n 26 Robotics ...
n n n
VII Conclusions
n
n n n n 27 Philosophy, Ethics, and Safety of AI ...
n n n n n n n n
n n n n 28 The Future of AI
n n n n
, EXERCISES n n
1
INTRODUCTION
Note nthat nfor nmany nof nthe nquestions nin nthis nchapter, nwe ngive nreferences nwhere
nanswers ncan nbe nfound nrather nthan nwriting nthem nout—the nfull nanswers nwould
nbe nfar ntoo n long.
1.1 n What Is AI?
n n
Exercise n1.1.#DEFA
Define nin nyour nown nwords: n (a) nintelligence, n(b) nartificial nintelligence, n(c) nagent, n(d)
nra- ntionality, n(e) nlogical nreasoning.
a. Dictionary ndefinitions nof nintelligence ntalk nabout n“the ncapacity nto nacquire
nand napply nknowledge” nor n“the nfaculty nof nthought nand nreason” nor n“the
nability nto n comprehend nand nprofit nfrom nexperience.” n These nare nall
nreasonable nanswers, nbut nif nwe nwant nsomething nquantifiable nwe nwould
nuse nsomething nlike n“the nability nto nact nsuccessfully nacross na nwide nrange nof
nobjectives nin ncomplex nenvironments.”
b. We ndefine nartificial nintelligence nas nthe nstudy nand nconstruction nof nagent
nprograms nthat nperform nwell nin na ngiven nclass nof nenvironments, nfor na
ngiven nagent narchitecture; nthey ndo nthe nright nthing. n An nimportant npart nof
nthat nis ndealing nwith nthe nuncertainty nof nwhat n the ncurrent nstate nis, nwhat
nthe noutcome nof npossible nactions nmight nbe, nand n what nis nit nthat nwe nreally
ndesire.
c. We ndefine nan nagent nas nan nentity nthat ntakes naction nin nresponse nto npercepts
nfrom nan nenvi- nronment.
d. We ndefine nrationality nas nthe nproperty nof na nsystem nwhich ndoes nthe n“right
nthing” ngiven nwhat nit nknows. n See nSection n 2.2 nfor na nmore n complete
ndiscussion. n The nbasic nconcept nis nperfect nrationality; nSection n?? ndescribes
nthe nimpossibility nof nachieving nperfect nrational- nity nand nproposes nan
nalternative ndefinition.
e. We ndefine nlogical nreasoning nas nthe na nprocess nof nderiving nnew nsentences
nfrom nold, nsuch nthat nthe nnew nsentences nare nnecessarily ntrue nif nthe nold nones
nare ntrue. n(Notice nthat ndoes nnot nrefer nto nany nspecific nsyntax nor nformal
nlanguage, nbut nit ndoes nrequire na nwell-defined nnotion nof ntruth.)
© n2023 nPearson nEducation, nHoboken, nNJ. nAll nrights
nreserved.