Artificial Intelligence: A Modern Approach, 4th Edition
by Peter Norvig and Stuart Russell, Chapters 1 – 28
,Artificial Intelligence
v
1 Introduction ...
vvvv v v v
2 Intelligent Agents ...
vvvv v v v v
II Problem-solving
v
3 Solving Problems by Searching ...
vvvv v v v v v v
4 Search in Complex Environments ...
vvvv v v v v v v
5 Adversarial Search and Games ...
vvvv v v v v v v
6 Constraint Satisfaction Problems ...
vvvv v v v v v
III Knowledge, reasoning, and planning
v v v v
7 Logical Agents ...
vvvv v v v v
8 First-Order Logic ...
vvvv v v v v
9 Inference in First-Order Logic ...
vvvv v v v v v
10 Knowledge Representation ...
vvvv v v v v
11 Automated Planning ...
vvvv v v v v
IV Uncertain knowledge and reasoning
v v v v
12 Quantifying Uncertainty ...
vvvv v v v v
13 Probabilistic Reasoning ...
vvvv v v v v
14 Probabilistic Reasoning over Time ...
vvvv v v v v v v
15 Probabilistic Programming ...
vvvv v v v v
16 Making Simple Decisions ...
vvvv v v v v v
17 Making Complex Decisions ...
vvvv v v v v v
18 Multiagent Decision Making ...
vvvv v v v v v
V Machine Learning
v v
, 19 Learning from Examples ...
vvvv v v v v v
20 Learning Probabilistic Models ...
vvvv v v v v v
21 Deep Learning ...
vvvv v v v v
22 Reinforcement Learning ...
vvvv v v v v
VI Communicating, perceiving, and acting
v v v v
23 Natural Language Processing ...
vvvv v v v v v
24 Deep Learning for Natural Language Processing ...
vvvv v v v v v v v v
25 Computer Vision ...
vvvv v v v v
26 Robotics ...
vvvv v v v
VII Conclusions
v
27 Philosophy, Ethics, and Safety of AI ...
vvvv v v v v v v v v
28 The Future of AI
vvvv v v v v
, EXERCISES v v
1
INTRODUCTION
Notevthatvforvmanyvofvthevquestionsvinvthisvchapter,vwevgivevreferencesvwherevanswersvcanvb
evfoundvrathervthanvwritingvthemvout—thevfullvanswersvwouldvbevfarvtoovlong.
1.1 What Is AI?
v v v
Exercisev1.1.#DEFA
Definevinvyourvownvwords:v (a)vintelligence,v(b)vartificialvintelligence,v(c)vagent,v(d)vra-
vtionality,v(e)vlogicalvreasoning.
a. Dictionaryvdefinitionsvofvintelligencevtalkvaboutv“thevcapacityvtovacquirevandvapplyvk
nowledge”vorv“thevfacultyvofvthoughtvandvreason”vorv“thevabilityvtovcomprehendvandv
profitvfromvexperience.”v Thesevarevallvreasonablevanswers,vbutvifvwevwantvsomethingv
quantifiablevwevwouldvusevsomethingvlikev“thevabilityvtovactvsuccessfullyvacrossvavwid
evrangevofvobjectivesvinvcomplexvenvironments.”
b. Wevdefinevartificialvintelligencevasvthevstudyvandvconstructionvofvagentvprogramsvtha
tvperformvwellvinvavgivenvclassvofvenvironments,vforvavgivenvagentvarchitecture;vtheyvd
ovthevrightvthing.v Anvimportantvpartvofvthatvisvdealingvwithvthevuncertaintyvofvwhatvth
evcurrentvstatevis,vwhatvthevoutcomevofvpossiblevactionsvmightvbe,vandvwhatvisvitvthatv
wevreallyvdesire.
c. Wevdefinevanvagentvasvanventityvthatvtakesvactionvinvresponsevtovperceptsvfromvanvenvi
-vronment.
d. Wevdefinevrationalityvasvthevpropertyvofvavsystemvwhichvdoesvthev“rightvthing”vgiven
vwhatvitvknows.v SeevSectionv2.2vforvavmorevcompletevdiscussion.v Thevbasicvconceptvi
svperfectvrationality;vSectionv??vdescribesvthevimpossibilityvofvachievingvperfectvration
al-vityvandvproposesvanvalternativevdefinition.
e. Wevdefinevlogicalvreasoningvasvthevavprocessvofvderivingvnewvsentencesvfromvold,vsuc
hvthatvthevnewvsentencesvarevnecessarilyvtruevifvthevoldvonesvarevtrue.v(Noticevthatvdoesv
notvrefervtovanyvspecificvsyntaxvorvformalvlanguage,vbutvitvdoesvrequirevavwell-
definedvnotionvofvtruth.)
Exercisev1.1.#TURI
ReadvTuring’svoriginalvpapervonvAIv(Turing,v1950).v Invthevpaper,vhevdiscussesvseveralvobj
ectionsvtovhisvproposedventerprisevandvhisvtestvforvintelligence.vWhichvobjectionsvstillvcarry
©v2023vPearsonvEducation,vHoboken,vNJ.vAllvrightsvreserve
d.