Guide
User Manual:
Open the PDF directly: View PDF 
.
Page Count: 4
lab_ml
Lazy Machine Learning
!"#$%&'(%)*+%),-./%01
!2345#6
Assignment Description
76%89:;%<=>%42"%?:<<%<#=(6%=>2"8$
@;:65%=%5(='9%=;%=%;8=8#%;'=A#
B#:6C2(A#D#68%<#=(6:65
E2?%82%8#=A9%=%A2D'"8#(%92?%82%<#=(6%82%?:6%89#%5=D#%2C%F:D
Checking Out The Code
G#8%89#%A2H#%:6%89#%";"=<%?=4I
J(2D%42"(%KL%)).%5:8%H:(#A82(4+%("6%89#%C2<<2?:65%26%MNL$
git fetch release
git merge release/lab_ml -m "Merging initial lab_ml files"
7C%42"O(#%26%42"(%2?6%D=A9:6#+%42"%D=4%6##H%82%("6$
git fetch release
git merge --allow-unrelated-histories release/lab_ml -m "Merging initial lab_ml files"
76%89:;%<=>+%42"%?:<<%D=P#%";#%2C%89#%Graph%A<=;;I
MH5#;%=(#%H:(#A8#HI%&HH:65%insertEdge(u, v)%A(#=8#;%H:(#A8#H%#H5#%u -> vI
L##%89#%!2345#6%C2(%89:;%<=>%Q2(%A9#AP%2"8%89#%C:<#%graph.hRI
The Game of Nim
S9#%G=D#%2C%F:D%:;%=%;:D'<#%8?2%'<=4#(%5=D#%?:89%26<4%=%C#?%("<#;$
M=A9%5=D#%;8=(8;%?:89%k%82P#6;%26%89#%8=><#
L8=(8:65%?:89%0<=4#(%*+%'<=4#(;%=<8#(6=8#%8"(6;$
M=A9%8"(6+%=%'<=4#(%D=4%':AP%"'%*%2(%)%82P#6;
S9#%'<=4#(%?92%':AP;%"'%89#%<=;8%82P#6%Q2(%82P#6;R%?:6;
T4%H#C="<8+%89#%5=D#%";"=<<4%;8=(8;%?:89%*U%82P#6;%Qk%VV%*URI%W2"%;92"<H%'<=4%=%C#?%(2"6H;%?:89%42"(%<=>
'=(86#(%82%5#8%=6%"6H#(;8=6H:65%2C%89#%5=D#I
Part 1: A State Space for Nim
&%state space%:;%=%D=89#D=8:A=<%(#'(#;#68=8:26%2C%89#%;8=8#%2C%=%'94;:A=<%;4;8#DI%76%2"(%A=;#+%89#%'94;:A=<
;4;8#D%:;%89#%5=D#%2C%F:DI
Assignment
Description
Checking Out
The Code
The Game of
Nim
Putting it
Together
Grading
Information
Good luck!
&8%89#%;8=(8%2C%F:DQ*UR+%:8%:;%0<=4#(%*O;%8"(6%?:89%*U%82P#6;%=X=:<=><#I%N#%A=6%<=>#<%89:;%;8=8#%p1-10+%C2(%“Player 1
with 10 tokens available”.%N9#6%0<=4#(%*%8=P#;%=%82P#6+%89#%;8=8#%:;%62?%p2-9%Q2(+%:C%0<=4#(%*%8=P#;%8?2%82P#6;+
p2-8RI
M=A9%;8=8#%:;%=%X#(8#3%26%=%5(='9%?:89%H:(#A8#H%#H5#;%A266#A8:65%X#(8:A#;%?9#6%89#(#%:;%=%X=<:H%D2X#%>#8?##6
89#%8?2%;8=8#;I%S9#(#C2(#+%89#%;8=8#%;'=A#%?:89%=<<%89#%X=<:H%D2X#;%=;%#H5#;%D=P#;%89#%C2<<2?:65%5(='9$
Programming NimLearnerʼs Constructor
K2D'<#8#%89#%NimLearner%A26;8("A82(+%?9:A9%A(#=8#;%89#%X#(8:A#;%=6H%#H5#;%C2(%89#%;8=8#%;'=A#%2C%=%5=D#%2C
F:DI%76%A(#=8:65%89:;%5(='9+%D=P#%;"(#%82%A(#=8#%=%?#:598#H%5(='9%?9#(#%=<<%#H5#%?#:598;%=(#%:6:8:=<<4%;#8%82
Y#(2%QURI%N#%?:<<%";#%89#;#%#H5#%?#:598;%<=8#(%C2(%(#:6C2(A#D#68%<#=(6:65I
&%C#?%8#;8%A=;#;%9=X#%>##6%'(2X:H#H%C2(%42"%89=8%42"%A=6%("6%";:65$
./test [part=1]
Part 2: Creating a Series of Choices
Z6#%2C%89#%D2;8%A<=;;:A%D#892H;%2C%D=A9:6#%<#=(6:65%:;%reinforcement learningI%76%(#:6C2(A#D#68%<#=(6:65+
=6%=<52(:89D%:;%(#?=(H#H%C2(%D=P:65%=%522H%A92:A#%=6H%'"6:;9#H%C2(%D=P:65%=%'22(%A92:A#I%&C8#(%;2D#
8(=:6:65+%89#%=<52(:89D%?:<<%9=X#%=%>#88#(%:H#=%?9=8%A92:A#;%=(#%522H%=6H%?9=8%A92:A#;%=(#%'22(I
S2%=''<4%(#:6C2(A#D#68%<#=(6:65+%?#%6##H%=%;#(:#;%2C%A92:A#;I%76%89:;%<=>+%2"(%=<52(:89D%?:<<%<#=(6%F:D%89(2"59
#3=D:6:65%(=6H2D%'=89;%89(2"59%89#%5=D#I%J2(%#3=D'<#+%26#%(=6H2D%'=89%F:DQ*UR%:;%89#%C2<<2?:65$
p1-10 ➜ p2-8 ➜ p1-7 ➜ p2-6 ➜ p1-4 ➜ p2-3 ➜ p1-1 ➜ p2-0
[:;"=<<4$
!%Programming Tips%&%C#?%8:';%89=8%D=4%>#%";#C"<%H"(:65%'(25(=DD:65%89:;%'=(8$
W2"%?:<<%6##H%82%D2H:C4\A26;8("A8%;8(:65;I%S9#(#%=(#%8?2%'2'"<=(%?=4;%82%H2%89:;%:6%K]]$
B#D#D>#(%89=8%std::string%9=;%=6%2X#(<2=H#H%operator+I%W2"%A=6%^'<";_%8?2%;8(:65;%82
A26A=8#6=8#%89#%;8(:65;%825#89#(I%7C%42"%6##H%82%A26X#(8%=6%int%82%=%std::string+
std::to_string(i)%?:<<%(#8"(6%=%;8(:65%C2(%=6%:68%iI
&<8#(6=8:X#<4+%=%std::stringstream ss%=<<2?;%42"%82%>":<H%=%;8(:65%?:89%89#%;8(#=D
:68#(C=A#%";:65%89#%operator<<%Q`";8%<:P#%coutRI%J2(%#3=D'<#+%ss<<"p"<<player<<"t"
<<tokens;%=6H%ss.str()%A=6%>#%";#H%82%>":<H%=%;8(:65I
S9:;%5(='9%:;%=%H:(#A8#H%=A4A<:A%5(='9I%W2"%A=6%C:6H%=6%2(H#(%?9#(#%42"%A(#=8#%89#%X#(8:A#;%>#C2(#
42"%6##H%82%A(#=8#%89#%#H5#;%>#8?##6%89#DI%Which vertex has an out-degree of zero?
Programming playRandomGame
K2D'#<8#%NimLearner::playRandomGame+%?9:A9%(#8"(6;%=%(=6H2D%'=89%89(2"59%89#%5(='9%2C%89#%;8=8#%;'=A#
=;%=%vector<Edge>I
&%C#?%8#;8%A=;#;%9=X#%>##6%'(2X:H#H%C2(%42"%89=8%42"%A=6%("6%";:65$
./test [part=2]
Part 3: Teaching a Machine to Learn
J:6=<<4+%?#%6##H%82%=''<4%(#:6C2(A#D#68%<#=(6:65I%G:X#6%=%'=89+%?#%?=68%82%(#?=(H%522H%H#A:;:26;I%N#%?:<<
H#C:6#%=%522H%H#A:;:26%=;%all decisions%D=H#%>4%89#%'<=4#(%?92%?26I%S9#(#C2(#+%:C%0<=4#(%*%822P%89#%<=;8%82P#6
Q=;%:6%2"(%#3=D'<#R+%=<<%A92:A#;%D=H#%>4%0<=4#(%*%:;%(#?=(H#HI
S9#%(#?=(H%:;%A='8"(#H%:6%2"(%=<52(:89D%=;%89#%#H5#%?#:598I%N9#6%?#%A26;:H#(%=%'=89%89(2"59%89#%5(='9+%?#
A=6%C:6H%89#%=<<%#H5#;%=<265%=%'=89%89=8%9=;%0<=4#(%*%?:66:65%Q#5$%89#%<=;8%X#(8#3%:6%89#%'=89%52#;%82%0<=4#(%)
?:89%62%82P#6;%(#D=:6:65+%2(%^')aU_+%D#=6:65%89=8%0<=4#(%*%822P%89#%<=;8%82P#6R+%89#6%=<<%A92:A#;%D=H#%>4
0<=4#(%*%Q#H5#;%?9#(#%0<=4#(%*%:;%89#%;2"(A#%X#(8#3R%=(#%(#?=(H#H%>4%:6A(#=;:65%89#%#H5#%?#:598%>4%]*%=6H%=<<
A92:A#;%D=H#%>4%0<=4#(%)%=(#%'"6:;9#H%>4%A9=65:65%89#%#H5#%?#:598%>4%a*I
&C8#(%26#%"'H=8#+%;#X#(=<%#H5#;%?:<<%9=X#%=%?#:598%2C%1+%;#X#(=<%#H5#;%?:<<%9=X#%=%?#:598%2C%-1+%892"59%D2;8
#H5#;%9=X#%628%>##6%X:;:8#H%=6H%?:<<%9=X#%89#:(%:6:8:=<%?#:598%2C%0I%&C8#(%;#X#(=<%892";=6H%"'H=8#;%2C%(=6H2D
5=D#;+%#H5#;%89=8%(#;"<8%:6%D2(#%X:A82(:#;%?:<<%9=X#%:6A(#=;:65<4%<=(5#%#H5#%?#:598;I
Programming updateEdgeWeights
K2D'<#8#%NimLearner::updateEdgeWeights+%?9:A9%"'H=8#;%89#%#H5#%?#:598;%=<265%=%5:X#6%path%26%89#%5(='9
2C%89#%;8=8#%;'=A#I
Putting it Together
&%main.cpp%:;%'(2X:H#H%89=8%'<=4;%*U+UUU%(=6H2D%5=D#;%2C%F:DQ*UR%=6H%'(2X:H#;%=6%2"8'"8%2C%89#%;8=8#%;'=A#
=8%89#%#6HI%W2"%A=6%("6%89:;%5=D#%>4$
./lab_ml
b22P:65%=8%89#%(#;"<8;+%42"%;92"<H%>#%=><#%82%=6;?#(%;#X#(=<%c"#;8:26;$
N9#6%89#(#O;%#3=A8<4%8?2%82P#6;%<#C8+%?9=8%;92"<H%0<=4#(%*%H2d%N9=8%=(#%89#%#H5#%?#:598;%2C%89#%2"8
!%Programming Tips%&%C#?%8:';%89=8%D=4%>#%";#C"<%H"(:65%'(25(=DD:65%89:;%'=(8$
S9#%rand()%C"6A8:26%(#8"(6;%=%(=6H2D%:68#5#(%>#8?##6%0%=6H%RAND_MAX%Q=%X#(4%<=(5#%6"D>#(RI%78O;
A2DD26%82%";#%rand() % size%82%5#8%=%(=6H2D%:68#5#(%:6%89#%(=65#%e0+%size - 1fI
MX#(4%'=89%D";8%;8=(8%2"8%=8%p1-#+%?9#(#%-#%:;%89#%6"D>#(%2C%;8=(8:65%82P#6;+%=6H%D";8%#6H%=8
p_-0+%?9#(#%p_%:;%89#%<2;:65%'<=4#(I
!%Programming Tips%&%C#?%8:';%89=8%D=4%>#%";#C"<%H"(:65%'(25(=DD:65%89:;%'=(8$
S9#%Graph%A<=;;%(#8"(6;%=%A2'4%2C%89#%Edge%;8("A8"(#+%628%89#%2(:5:6=<%26#g%S9#(#C2(#+%"'H=8:65%89#
Edge::weight%'(2'#(84%2C%89#%#H5#%H2#;%628%"'H=8#%89#%#H5#%?#:598%Q#5$%42"%A=6%89:6P%2C%:8%=;
(#=Ha26<4RI
76;8#=H+%Graph::setEdgeWeight%;92"<H%>#%";#H%82%"'H=8#%89#%#H5#%?#:598%2C%=%5:X#6%#H5#I
#H5#;%2C%p1-2%=6H%p2-2d
N9#6%89#(#%=(#%C2"(%82P#6;%<#C8+%?9=8%A=6%0<=4#(%)%H2d%N9=8%=(#%89#%#H5#%?#:598;%2"8%2C%p1-4%=6H%p2-4d
7;%p1-10%2(%p2-9%=%>#88#(%;'28%82%>#%:6d
N2"<H%42"%'(#C#(%82%52%C:(;8%2(%;#A26H%:6%F:DQ*URd
Grading Information
S9#%C2<<2?:65%C:<#;%=(#%";#H%C2(%5(=H:65%89:;%<=>$
NimLearner.h
NimLearner.cpp
7C%42"%D2H:C4%=64%289#(%C:<#;+%89#4%?:<<%628%>#%5(=>>#H%C2(%5(=H:65%=6H%42"%D=4%#6H%"'%?:89%=%^;8"':H%Y#(2I_
"%G":H#$%E2?%82%;">D:8%KL%)).%?2(P%";:65%5:8
Good luck!
