Lecturer: ¦¿®¶·ç
Teaching Assistants (TAs):
³¯¾ÇP §õ¨ÎÀM ¹ù«Ø³Í ³¯ªu§Ê
Time: ¶g¤G 14:00~16:50
Place: ¶g¤G ¤u¤À]E6-A207
TA Class:
¶g¤G 13:00~13:50
(¤u¤À]E6-A207)
Online
Video: https://www.twitch.tv/acanlab
Offline Video: https://drive.google.com/folderview?id=1WsZopoti_c_7gKfFHerIk5iflLGsQc4_
Scoring¡G
- ´Á¤¤¦Ò(¤ÀªR»P³]p·§©À)(20%)
- ´Á¥½¦Ò(¤ÀªR»P³]p·§©À)(30%)
- µ{¦¡³]p§@·~¡B³ø§i¡B½Òµ{°Ñ»P«×¤Îµ{¦¡³]pÄvÁɦ¨ÁZ(50%)
Textbooks:
- My Book's Manuscript: book1-7AB.zip
(°É»~: page 39: ¨BÆJ5:
j <- 2i+1 §ï¬° j <- 2i¡C¡C
¹Ï3.4: 8'ªº¸`ÂI½s¸¹À³¬°#3)
- R.C.T. Lee et. al., Introduction to the Design and Analysis
of Algorithms -- A Strategic Approach (2/e), McGraw Hill,
2005.
- Thomas Cormen, Charles
Leiserson, Ronald Rivest and Clifford Stein, Introduction to Algorithms (3/e),
MIT Press, 2009.
Reference Books:
- ªL¤j¶Q, TensorFlow + Keras ²`«×¾Ç²ß¤H¤u´¼¼z¹ê°ÈÀ³¥Î, ³ÕºÓ¤å¤Æ, 2017.
- »ôÃñd¼Ý, Deep Learning -- ¥ÎPython¶i¦æ²`«×¾Ç²ßªº°ò¦²z½×¹ê§@, O'REILLY, (§d¹ÅªÚ
Ķ, ùÖ®p¸ê°T¥Xª©), 2017.
- Algorithm Design: Foundations, Analysis, and Internet
Examples, Michael T. Goodrich, Roberto Tamassia, Wiley, 2002.
(¦³¤¤Ä¶¥»)
- Computer Algorithms, Ellis Horowitz et. al., Silicon Press,
2008.
-
¤¤¤åJavaµ{¦¡³]p (²Ä¤Tª©) SIM-908A , ¦¿®¶·ç µÛ, ¾§ªL¥Xª©ªÀ, 2006.
-
ª«¥ó¾É¦V¸ê®Æµ²ºc ¡X ¨Ï¥ÎJava»y¨¥, ¦¿®¶·ç µÛ, ªQ±^¹Ï®Ñ¤½¥q, 2005.
- Algorithms, S. Dasgupta, C. Papadimitriou and U. Vazirani,
McGraw Hill, 2008.
- Programming Challenges, Steven S. Skiena and Miguel Revilla,
Elsevier, 2003.
- The Algorithm Design Manual, Steven S. Skiena, Elsevier.
Programming:
¦w¸Ë½Òµ{¬ÛÃö³nÅé¤Îì©lµ{¦¡½X: Jeep7 (for
JAVA platform on Windows)
1 ¦w¸ËJava Development Kit (JDK): http://java.sun.com/javase/downloads/index.jsp
2 ¦w¸ËMinGW C++ Compiler
(G++):
http://prdownloads.sf.net/mingw/MinGW-3.0.0-1.exe?download
3 ¦w¸Ë³Ì·sPython 3 Interpreter:
https://www.anaconda.com/
4 ¦w¸ËJeep7(
Java
editor for
every
programmers v7.0)
(¤U¸ü: Jeep7Setup&SourceCode.exe)
(¨¾¬r³nÅé·|»~§P¬°¤£¦w¥þ³nÅé¡A¦ý«OÃÒ¦w¥þ¡A½Ð¦w¤ß¨Ï¥Î¡C)
5 ³]©w¸ô®|°Ñ¼Æ:
ª`·N¡A³o·|¦]¬°§A¦w¸Ë¤W¦C¤£¦P³nÅ骺¤£¦Pª©¥»¦Ó¤£¦P¡A¤]·|¦]¬°§A¦w¸Ëªº§@·~¨t²Îª©¥»¤£¦P¦Ó¤£¦P¡C¥H¤U¬°°w¹ïWindows§@·~¨t²Îªº³]©w:
¿ï¾Ü
[±±¨î¥x][¨t²Î¤Î¦w¥þ©Ê][¨t²Î][¶i¶¥¨t²Î³]©w][Àô¹ÒÅܼÆ]¡A§ä¥X[path]Åܼƨëö¤U[½s¿è]¡A¨Ã¦b¨äÅܼÆÈ¥½ºÝ¥[¤J¥H¤U¤º®e:
;JDK¦w¸Ë¥Ø
¿ý\bin;MinGW¦w¸Ë¥Ø¿ý\ bin;Python¦w¸Ë¥Ø¿ý;Jeep7¦w¸Ë¥Ø¿ý
(¨Ò¦p: ;C:\Program
Files\Java\jdk-9.0.4\bin;C:\MinGW\bin;C:\Users\yourname\Anaconda3;C:\Jeep7)
6 ¤U¸ü½d¨Òµ{¦¡(
Sample.c)(
Sample.cpp)(
Sample.java)(
Sample.py)Àx
¦s©óC:\Jeep7¥Ø¿ý¤¤¡A«ö¤U®à±Jeep7¹Ï¥Ü°õ¦æJeep7³nÅé¡A¨Ã¨Ï¥Î [¶}ÂÂÀÉ]¿ï¶µ¸ü¤J½d¨Òµ{¦¡¶i¦æ[½sĶ][°õ¦æ]¡C
ACM°ê»Ú¤j¾Ç¥Í
µ{¦¡³]p ÄvÁÉ (ACM International Collegiate Programming Contest,
ACM-ICPC) (ACM-ICPC&EPC.ppt)
Syllabus:
- 1. »{ÃѺtºâªk -- ±q¹ÃШ찪¶¥µ{¦¡»y¨¥:
(AlgSmallTalk.pptx) (Alg-Intro.pptx) (CPSforIndustry4.0.zip) (ACM-ICPC&EPC.ppt)(CPSProject.zip)(3/3
and 3/10)
1. ºtºâªk¦WºÙªº¥Ñ¨Ó
2. ¤°»ò¬Oºtºâªk?
3. ºtºâªkªº¨Ò¤l
4. ¦p¦óªí¥Üºtºâªk?
5. ¦p¦ó¹ê§@ºtºâªk? (EuclidGCD.c)(EuclidGCD.cpp)(EucidCGDClass1.java)(EucidCGDClass.java)(EuclidGCD.py)
6. ºtºâªkªº¥¿½T©Ê
Homework1: (for 3/10; Due day:
before next class or TA's class)
(A)¨Ï¥ÎµêÀÀº¿(pseudo
code)¼g¤@Óºtºâªk¡A¿é¤J¤@Ó¾ã¼Æn(n>2)¨Ã¨Ã§PÂ_n¬O§_¬°½è¼Æ(prime)(Write an
algorithm using the pseudo code to input an integer n and
output (decide) if n is a prime.)
(B) ¨Ï¥ÎµêÀÀ º¿(pseudo
code)¼g¤@Óºtºâªk¡A¿é¤J¤@Ó¾ã¼Æn(n>2)¨Ã¿é¥X¤p©ónªº³Ì¤j¦]¼Æ(factor) (Write an
algorithm using the pseudo code to input an integer n and
output the n's largest factor that is less than n.)
(C) ¨Ï¥ÎµêÀÀ
º¿(pseudo code)¼g¤@Óºtºâªk¡A¿é¤J¤@Ó¾ã¼Æn(n>2)¨Ã¿é¥X©Ò¦³n°£¤F¥»¨¥H¥~ªº¥¿¦]¼Æ(factor)Á`©M
(Write an algorithm using the pseudo code to input an integer
n and output the total summation of all n's factors except n.)
(D) ¨Ï¥ÎµêÀÀº¿(pseudo
code)¼g¤@Óºtºâªk¡A¿é¤J¾ã¼Æn¤Îm(n>m>2)¡A¿é¥X©Ò¦³¤ñn¤p¨Ã¤j©ómªºnªº¦]¼Æ(factor)Á`©M¡AYµL
¦¹¦]¼Æ«h¿é¥X0 (Write an algorithm using the pseudo code to input
integers n and m, and output all n's factors larger than m and
less than n.)
(E) ¨Ï¥ÎµêÀÀº¿(pseudo code)¼g¤@Óºtºâªk¡A¿é¤J¤@Ó¾ã¼Æn(n>2)¨Ã§PÂ_n¬O§_¬°§¹¬ü¼Æ(perfect
number)¡C¤@Ó§¹¬ü¼Æ¬O¤@Ó¥¿¾ã¼Æ¡A¥¦©Ò¦³ªº¯u¦]¼Æ(§Y°£¤F¦Û¨¥H¥~ªº¦]¼Æ)ªº©M¡A«ê¦nµ¥©ó¥¦¥»¨¡C (Write an
algorithm using the pseudo code to input an integer n and
output (decide) if n is a perfect number. Note that a perfect
number is a positive integer that is equal to the sum of its
proper positive divisors, that is, the sum of its positive
divisors excluding the number itself.)
(F) ¨Ï¥ÎµêÀÀº¿(pseudo code)¼g ¤@Óºtºâªk¡A¿é¤J¤@Ó¾ã¼Æn(n>0)¨Ã§PÂ_n¬O§_¬°§Ö¼Ö¼Æ(happy
number) (Write an algorithm to input an integer n and output
(decide) if n is a happy number.)
µù:
§Ö¼Ö¼Æ¦³¥H¤Uªº¯S©Ê¡G¦bµ¹©wªº¶i¦ì¨î¤U¡A¸Ó¼Æ¦r©Ò¦³¼Æ¦ì(digits)ªº¥¤è©M¡A±o¨ìªº·s¼Æ¦A¦¸¨D©Ò¦³¼Æ¦ìªº¥¤è©M¡A¦p¦¹«½Æ¶i
¦æ¡A³Ì²×µ²ªG¥²¬°1¡C¨Ò ¦p¡A¥H¤Q¶i¦ì¬°¨Ò¡G
28 ¡÷ 22+82= 68 ¡÷ 62+82=100
¡÷
12+02+02=1¡A¦]¦¹28¬O§Ö¼Ö¼Æ
- 7. ³g°ýºtºâªk(greedy algorithm):
(Alg-Greedy.pptx) (4/21)
³g°ýºtºâªk(greedy
algorithm)¤@¨B¨B¦a«Øºc¥X¤@Ó°ÝÃDªº§¹¾ã¸Ñµª¡C¨ä¨C¤@¨B³£Âǥѳg°ý¸ÑÃDµ¦²¤(greed
strategy)¼W¥[¤@³¡¥÷ªº¸Ñµª¨ì§¹¾ã¸Ñµª¤¤¡C©Ò¿×³g°ý¸ÑÃDµ¦²¤¬°:
¨C¤@¦¸³£¿ï¾Ü·í¤U³Ì¦nªº³¡¥÷¸Ñµª¥[¤J§¹¾ã¸Ñµª¤¤¡C
.1 I¥]ºtºâªk(knapsack algorithm)
(±a¥X0/1I¥]°ÝÃD«á¥i¥H¦b¤§«áÁ¿±ÂNP-hard°ÝÃD)
.2 ¬¡°Ê¿ï¾Ü(activity
selection)ºtºâªk
.3 Huffman½s½X(Huffman coding)ºtºâªk
.4 Kruskal³Ì¤p¥Í¦¨¾ð(minimum spanning
tree, MST)ºtºâªk
.5 Prim³Ì¤p¥Í¦¨¾ð(minimum spanning
tree, MST)ºtºâªk
Homework 7:
(A) ¤@ÓI¥]®e¶q¬°
10¡A²{¦b¦³5Óª««~¡A «¶q¤À§O¬°4¡B3¡B6¡B2¡B5¡A»ù®æ¤À§O¬°10¡B9¡B
12¡B4¡B8¡A¨DI¥]¯à°÷¸Ë¤J¹s¸H(fractional)ª««~ªº³Ì¤j»ùȬ°¦ó?
(B) µ¹©w 5 Ó¬¡°Ê¡A¨ä¬¡°Ê°Ï¶¡¤À§O¬° [0,18), [3, 5), [4, 16), [2, 9), [10,
15)¡A½Ð´yz¦p¦ó¥H¬¡°Ê¿ï¾Üºtºâªk§ä¥X³Ì¦hªº¬Û®e¬¡°ÊÁ`¼Æm(¥²¶·¼g¥Xmªº¼ÆÈ)¡C
(C)
§Q¥ÎHuffman ½s ½Xºtºâªk´À¥H¤U¦r¤¸½s½XA(46%)¡B B(7%)¡BC
(28%)¡BD (19%)¡C(³Æµù:¬A¸¹¤¤¬°¦r¤¸¥X²{ÀW²v)
(D) §Q¥ÎKruskalºtºâªk¨D¥X¥H¤Uªº¹Ï(graph)ªº³Ì¤p¥Í¦¨¾ð (minimum spanning tree, MST)
(E) §Q¥ÎPrimºtºâªk¨D¥X¥H¤Wªº¹Ï(graph)ªº³Ì¤p¥Í¦¨¾ð (minimum spanning tree,
MST)
- ´Á¤¤¦Ò: (®É
¶¡: 4/28 ¤U ¤È2:00-3:50)(½d³ò: ¤w±Â½Òªº³¡¥÷)(¦aÂI:
½ÐTA¥t¦æ¤½§G«ö·Ó®y¦ìªí¤J®y¡AÀ³¸Õ½ÐÀ¹¤f¸n)
±Ð§÷: book1-7AB.zip(´Á
¤¤¦Ò¸Õµª®×¥H¦¹¬°·Ç)(¾ð·j´M»P¦^·¹ºtºâªkªº¼Ð·Çµª®×«h¥H¸Ó½Òµ{ªº§ë
¼v¤ù¬°·Ç)
(±Ð§÷°É»~: page 39: ¨BÆJ5: j
<- 2i+1 §ï¬° j <- 2i¡C¹Ï3.4: 8'ªº¸`ÂI½s¸¹À³¬°#3)
- Term Project »¡©ú»P±Ð¾Ç (5/5)
(¥»©PµL½Òµ{§@·~¤]µLµ{¦¡³]p§@·~¡A½Ð¦P¾Ç°È¥²§â´¤®É¶¡Æ[¬Ý¥Ñ§U±Ð¿ý»sªºTerm
Project±Ð¾Ç¼v¤ù¡A¾¨¦§¹¦¨¾÷¾¹¾Ç²ß/²`«×¾Ç²ßÀô¹Ò¦w¸Ë¡A¨Ã¶}©l¼¶¼gµ{¦¡¥H§¹¦¨Term Project)
ÃD¥Ø: ²`«×¾Ç²ß¤¤¤å¤â¼g¼Æ¦r¿ëÃÑ
¿é¤J: ¤¤¤å¤â¼g¼Æ¦r¸ê®Æ¶°(dataset)
¿é¥X: ¤¤¤å¤â¼g¼Æ¦r¿ëÃѷǽT²v(accuracy)
¦û¤À: Á`¦¨ÁZ20%
¥Øªº: Åý¾Ç¥Í½m²ß¨Ï¥Î²`«×¾Ç²ß(deep learning)¼Ò«¬¸Ñ¨M¤¤¤å¤â¼g¼Æ¦r¿ëÃÑ°ÝÃD¡C²`«×¾Ç²ß(deep learning)¼Ò«¬¥i¬°²`«×¯«¸gºô¸ô(deep neural network, DNN)¡B±²¿n¯«¸gºô¸ô(convolutional neural network, CNN)¡Bªøµu´Á°O¾Ð(long short-term memory, LSTM)¯«¸gºô¸ô©Î¹h±±»¼°j³æ¤¸(gated recurrent unit, GRU)¯«¸gºô¸ô¡C
¸ê®Æ¶°»¡©ú: (¸ê®Æ¶°¤U¸ü³sµ²)
¸ê®Æ¶°ÀÉ®×handwrite_detect.zip¬°¤¤¤å¤â¼g¼Æ¦r¸ê®Æ¶°¡A¥]§t¤F°V½m¸ê®Æ¶°(training dataset)»P´ú¸Õ¸ê®Æ¶°(test dataset)
°V½m¸ê®Æ¦@¦³2450µ§
´ú¸Õ¸ê®Æ¦@¦³1700µ§
µû¤À¶µ¥Ø»P³W½d:
Source code(.py)
A short report (.pdf) (§t¦³)
¤¤¤å¤â¼g¿ëÃѷǽT²v(accuracy)¡A¥HºI¹Ï¤è¦¡§e²{
Source code¤§³v¦æ¸ÑÄÀ
ú¥æ¤è¦¡: ½Ð±N©Ò¦³ÀÉ®×À£ÁY¦¨ .zip «á¡A¦A¤W¶Ç¦Ü LMS §@·~ú¥æ°Ï
(¤W¶ÇÀɦW:Term_Project_¾Ç¸¹_©m¦W.zip)
ú¥æ®É¶¡: ´Á¥½¦Ò¨â©P«á©P¤G 20:00«eºI¤î
³Æµù : ¤Á¤Å¤¬¬Û§Ûŧ¡AYµo²{§Ûŧ¡A«h¥H0¤Àpºâ
§ë¼v¤ù: https://drive.google.com/drive/folders/1OiRiTyf5xMbHxlUEg2LUbXT9Z9GqgLZu?usp=sharing
±Ð¾Ç¼v¤ù:
ºtºâªk½Òµ{_¾÷
¾¹¾Ç²ßÀô¹Ò¦w¸Ë¤§±Ð¾Ç¼v¤ù
ºtºâªk½Òµ{_Machine Learning_01_DNN
ºtºâªk½Òµ{_Machine Learning_02_CNN
ºtºâªk½Òµ{_Machine Learning_03_LSTM
ºtºâªk½Òµ{_Machine Learning_04_GRU
- 8. °ÊºA³W¹º(dynamic
programming)ºtºâªk: (Alg-DP.pptx) (5/12)
°ÊºA³W¹º(dynamic programming)ºtºâªkÄy¥Ñ±Nì°ÝÃD¤À¸Ñ¦¨¤@¨t¦C¤l°ÝÃD
(subproblems)¡A¨Ã¨Ì§Ç¸Ñ¨M¤l°ÝÃD¨Ó¸Ñ¨Mì°ÝÃD¡C¬°ÁקK¤@¦A¦a¸Ñ«½Æªº¤l°Ý
ÃD¡A¤@¥¹¸Ñ¥X¤l°ÝÃDªº¸Ñµª(solution)¡A§Y·|±N¨ä¦s¦bªí®æ(©Î°}¦C)¤¤¡C·í»Ýn¥Î¨ì¬Y¤@¤l°ÝÃDªº¸Ñµª®É¡A»P¨ä«·spºâ¨ä
¸Ñµª¡Aºtºâªk·|¨ú¦Ó¥N¤§¦a
±qªí®æ¤¤ª½±µ¨ú¥X¨ä¸Ñµª¥H¸`¬Ùpºâ®É¶¡¡A¬O¤@Ó¡u¥HªÅ¶¡´«¨ú®É¶¡¡vªººtºâªk¡C¤@ӰʺA³W¹ººtºâªk·|¥ý±q³Ì²³æªº¤l°ÝÃD¥ý¸Ñ°_¡A¨Ã
¥H¤@©wªºµ{§Ç«ùÄò¹B¦æª½¦Ü ¨D¥Xì°ÝÃD¸Ñµª¬°¤î¡C
³Ì¨Î¸Ñì«h(Principle of optimality):
°²³]¬°¤F¸Ñ¨M¤@Ó°ÝÃD¡A§ÚÌ¥²¶·§@¥X¤@¨t¦Cªº¨Mµ¦ D1, D2, ¡K,
Dn¡CY³o¤@¨t¦Cªº¨Mµ¦¬O³Ì¨Î¸Ñ¡A«h°w¹ï©ó«en-kÓ(©Î³Ì«án-kÓ)¨Mµ¦©Ò²£¥Íªºª¬ºA(¤l°ÝÃD)¦Ó¨¥¡A³Ì«áªºkÓ(©Î«ekÓ)¨M
µ¦(1<= k<=n)¥²©w¤]¬O³Ì¨Îªº¡C
*³Ìªø¦@¦P¤l§Ç¦C(longest common
subsequence, LCS or LCSS)ºtºâªk
*³Ì¤p½s¿è¦¨¥» (Minimum Edit Cost, MEC)ºtºâªk
*0/1I¥]°ÊºA³W¹ººtºâªk(0/1 knapsack dynamic programming
algorithm)
*¤l¶°¦X¥[Á`(subset sum)°ÊºA³W¹ººtºâªk
*(½Æ²ß)¦h¶µ¦¡¤Î°°¦h¶µ¦¡®É¶¡ºtºâªk
*³Ì¤j³sÄò¤l§Ç¦C©M(maximum contiguous subsequence sum, MCSS)°ÊºA³W¹ººtºâªk
Homework 8:
A. »¡©úX=ABCBA»PY=BDCA¨âÓ¦r¦êÂǥѰʺA³W¹ººt
ºâªk¨D¥X³Ìªø¦@¦P¤l§Ç¦Cªº¹Lµ{¡C
B. ³]p¤@Óºtºâªk¡A¥i¥H²£¥Í¤@Óµ¹©w¶°¦XSªº©Ò¦³¥i¯à¤l¶°¦X¡C
C. ³]p¤@Óºtºâªk¡A¥i¥H¿é¤Jªø«×¬°mªº§Ç¦CX¤Îªø
«×¬°nªº§Ç¦CY¡A¿é¥Xtrue©Î¬Ofalse¡A¤À§O
¥NªíX¬O©Î¤£¬OYªº¤l§Ç¦C¡C
D.
µ¹©w¤@Ó0/1I¥]°ÝÃD¦p¤U¡FI¥]²ü«W=12¡A¥B4Óª««~¨ä«¶q¦U¬°6¡B4¡B5¡B3¡A¨ä»ùȦU¬°20¡B30¡B40¡B10¡A»¡©úÂǥѰʺA³W¹ººtºâªk
¸Ñ¨M¦¹0/1I¥]°ÝÃDªº¹Lµ{¡C
E. ¥H¤l¶°¦X¥[Á`°ÊºA³W¹ººtºâªk¸Ñ¨M¥H¤U¤l¶°¦X¥[Á`°ÝÃD: µ¹©w¾ã¼Æ¶°¦XS={1, 2, 4,
7}¤Î¾ã¼Æc=10¡C
F.
קï¤l¶°¦X¥[Á`°ÊºA³W¹ººtºâªk¨Ï¨ä¶Ç¦^¥[Á`Ȭ°cªº¤l¶°¦XY¦¹¤l¶°¦X¦s¦b¡F§_«h¶Ç¦^ªÅ¶°¦X¡C
G. ¥OS = (-2, 1, -3, 4, -1, 2, 1, -5, 4), ¨Ï¥Î°ÊºA³W¹ººtºâªk¨D¥X³Ì¤j³sÄò«DªÅ¤l
§Ç¦C©M¡C
H. ³]p³Ì¤j³sÄò¥iªÅ¤l§Ç¦C©M°ÊºA³W¹ººtºâªk¡C
- 9.
¨Ï¥Î³g°ý(greedy)ºtºâªk¥H¤Î°ÊºA³W¹º(dynamic programming)ºtºâªk¸Ñ¨M³Ìµu¸ô®|°ÝÃD: (Alg-SP.pptx) (5/19)
(½Ð
¦P¾Çª`·N: ¬°¨Ï»¡©ú§ó²M·¡¡A¥»¦¸½Òµ{§ë¼v¤ù¦b½Òµ{¿ý¼v¤§«á¦³¨Çקï¡A¨Ò¦p¡A"t
°j°é"§ï¬°"t´`Àô"µ¥¡A¦b´Á¥½¦Ò¸Õ®É¥Hºô¶»¡©ú¤Î§ë¼v¤ùªº¤º®e¬°·Ç¡C)
¥Ñ¥[Åv¦³¦V¹Ï(weighted digraph)¤¤ªº¬YÓ³»ÂI©Î¸`ÂI (vertex or
node)v¨ì¹Ï¤¤ªº¥t¤@¸`ÂIu¡AYv¨ìu¤§¶¡¦s¦b¤@±ø¸ô®|(path)¡A«h¸ô®|¤¤©Ò¸g¹LªºÃä(edge)ªºÅvÈ(weight)Á`¦XºÙ¬°¸ô®|ªº¦¨¥»
(cost)©Î¶ZÂ÷(distance)¡A¦Ó©Ò¦³¸ô®|¤¤¨ã¦³³Ì¤p¦¨¥»©Î¶ZÂ÷ªº¸ô®|«hºÙ¬°³Ìµu¸ô®|(shortest
path)¡C
µÛ¦Wªº³Ìµu¸ô®|ºtºâªk¥]¬A¡G
(1) ¦h¶¥¹Ï³Ìµu¸ô®|ºtºâªk(¨Ï¥Î°ÊºA³W¹º¸ÑÃDµ¦²¤)
(2) Dijkstraºtºâªk(¨Ï¥Î³g°ý¸ÑÃDµ¦²¤)
(3) Bellman-Fordºtºâªk(¨Ï¥Î°ÊºA³W¹º¸ÑÃDµ¦²¤)
(4) Floyd-Warshallºtºâªk(¨Ï¥Î°ÊºA³W¹º¸ÑÃDµ¦²¤)
Homework 9:
A.
¨Ï¥Î°ÊºA³W¹ººtºâªk¨D¥H¤U¦h¶¥¹Ïªº³Ìµu¸ô®|¡C
B.
§Q¥ÎDijkstraºtºâªk¨D¥H¤U¹Ï(graph)³»ÂI4¨ì¦U³»ÂIªº³Ìµu¸ô®|(shortest
path)¤Î¨ä¶ZÂ÷(¦¨¥»)¡C
C. ©Ó¤WÃD¡A§Q¥ÎDijkstraºtºâªk¨D³»ÂI1¨ì¦U³»ÂIªº³Ìµu¸ô®|(shortest
path)¤Î¨ä¶ZÂ÷(¦¨¥»)¡C
D.
µe¹Ï»¡©ú§Q¥Î§Q¥ÎFloyd-Warshallºtºâªk¨D¥H¤U¹Ï(graph)¥þ¹ï³Ìµu¸ô®|(all-pair shortest
path)¶ZÂ÷(¦¨¥»)(¦¹¹Ïªº±Ò©l¶ZÂ÷¯x°}¦p¤U¡A¥H¸g¹Lªº¤¤¶¡¸`ÂI¬°s, a, b, c, dªº¶¶§Ç¼g¥X¶ZÂ÷¯x°}ªº§ïÅܹLµ{¡C)
(d->bªº¥[Åv¦b¹Ï§Î»Pªí®æ¤¤»~´Ó¬°¤£¤@PªºÈ¡A¦P¾Ç¥i¥H±N¤§§ï¬°5©Î6Åý¹Ï§Î»Pªí®æ¤@P«á¸ÑÃD³£ºâµª¹ï¡C)
E.
¨D¥X¥H¤Uµ¹©w¹Ï(graph)ªºFloyd-Warshallºtºâªkªº±Ò©l«e¸`ÂI¯x°}(°}¦C)¡A¨Ã¨D¥X³Ì«áªº«e¸`ÂI¯x°}(°}¦C)¡C
F.
¥H¤U¬OFloyd-Warshallºtºâªk°w¹ï¨ã¦³¤Ó¸`ÂI(°O¬°1¡B2¡B3¡B4¡B5)ªº¹Ï²£¥Íªº«e¸`ÂI¯x°}(°}¦C)¡A»¡©ú¦p¦óÂÇ¥H
§ä¥X¸`ÂI1¨ì¸`ÂI3ªº ³Ìµu¸ô®|¡A¤Î¸`ÂI5¨ì¸`ÂI2ªº³Ìµu¸ô®|¡C
G. °w¹ï¥H¤Uªºµ¹©w¹Ï¡A¦C¥XBellman-Ford³Ìµu¸ô®|ºtºâ ªk°õ¦æ¹Lµ{¡A »¡©úBellman
-Ford³Ìµu¸ô®|ºtºâªk¦p¦óÀˬd¥X¤@µ¹©w¹Ï¨ã¦³t´`Àô(negative-weight cycle)¡C
- Introduction to Software Defined
Networking (SDN) Multicast (SDN-Multicast.pptx)
(Optional)
- 10.
¤À¤ä©w¬É(Branch and
Bound)ºtºâªk (Branch&Bound.zip)
(Youtube video Part1: https://youtu.be/KUlDxRV6fsU)(Youtube video Part2: https://youtu.be/R91wC81n6Yk)
(5/26)
*¤À¤ä©w¬É(branch and bound)¬O¤@Ó«ô ³X»P©Ý®i¸Ñ
µªªÅ¶¡¾ðªº¯S®í¤èªk¡A¥Î©ó§ä¥X°ÝÃDªº³Ì¨Î¸Ñ¡C¨ä°µªk¬Û·í©ó§ä¥X¤@ºØ¤èªk¨Ó¤Á³Î¥X¸ÑµªªÅ¶¡ªº¤À¤ä(branch)¡AµM«á¥H¤W¬É
(upper bound)»P¤U¬É(lower bound)ªº·§©À¨Ó¥[§Ö³Ì¨Î¸Ñªº·j´M¡C
*¹ï©ó´M§ä³Ì¤p¦¨¥»(minimum
cost)¸Ñµªªº³Ì¨Î¤Æ°ÝÃD¦Ó¨¥¡A¤À¤ä©w¬Éºtºâªk·|°w¹ï¨C¤@¤À¤äªº¸Ñµª¹w´ú¨ä¦¨¥»¤U¬É(lower
bound)¡A¨Ã§Q¥Î§ä¥X¥i¦æ¸Ñ¨Ó±o¨ì°ÝÃDªº¦¨¥»¤W¬É(upper
bound)¡C¦pªG¦³¤@Ӹѵªªº¦¨¥»¤U¬É¶W¹L°ÝÃD¦¨¥»¤W¬É¡A«h³oӸѤ£¥i¯à¬O³Ì¨Îªº¡A¦]¦¹ºtºâªk·|¤¤¤î(terminate)»P
³oӸѬÛÃöÁpªº¾ãÓ¤À¤äªº ·j´M¡C
*¦Ó¹ï©ó´M§ä³Ì¤j§Q¯q(maximum
benefit)¸Ñµªªº³Ì¨Î¤Æ°ÝÃD¦Ó¨¥¡A¤À¤ä©w¬Éºtºâªk·|°w¹ï¨C¤@¤À¤äªº¸Ñµª¹w´ú¨ä§Q¯q¤W¬É(upper
bound)¡A¨Ã§Q¥Î§ä¥X¥i¦æ¸Ñ¨Ó±o¨ì°ÝÃDªº§Q¯q¤U¬É(lower
bound)¡C¦pªG¦³¤@Ӹѵªªº§Q¯q¤W¬É§C©ó°ÝÃD§Q¯q¤U¬É¡A«h³oӸѤ£¥i¯à¬O³Ì¨Îªº¡A¦]¦¹ºtºâªk·|¤¤¤î(terminate)»P
³oӸѬÛÃöÁpªº¾ãÓ¤À¤äªº ·j´M¡C
.1 °ò¥»·§©À
.2 ¤À¤ä©w¬Éºtºâªk(°t¦Xµn¤s·j´Mªk)¸Ñ¨M¦h¶¥¹Ï³Ìµu¸ô®|°ÝÃD
.3 ¤À¤ä©w¬Éºtºâªk(°t¦X³Ì¨ÎÀu¥ý·j´M
ªk)¸Ñ¨M®È¦æ±À¾Pû°ÝÃD
.4 ¤À¤ä©w¬Éºtºâªk(°t¦X³Ì¨ÎÀu¥ý·j´M
ªk)¸Ñ¨M0/1I¥]°ÝÃD
.5 ¤@Ó¯S®íªº¤À¤ä©w¬Éºtºâªk: A*ºtºâªk
Problem Set 10:
A.
¥H¤À¤ä©w¬Éºtºâªk¸Ñ¨M¥H¤U¦h¶¥¹Ï³Ìµu¸ô®|°ÝÃD¡C
B.
µ¹©w¤@0/1I¥]°ÝÃD¡A¨äI¥]¥i¸ü«C=10,¥B¦³¤Tª««~ªº«¶q¤À§O¬°10, 3, 5¡A¦Ó¨ä§Q¼í¤À§O¬°40, 20,
30¡A¨D¥X¥i¥H©ñ¤JI¥]¤¤ª««~§Q¼íªºtȪº³Ì¤p¤Æ¡A¨Ï¥Î¤À¤ä©w¬Éºtºâªk¨Ó¸Ñ¦¹0/1I¥]°ÝÃD¡C ¡]µù:
¥²¶·µe¥X·j´M¾ð¡^(¦¹ÃD¨Ï¥Îtªº¥Ø¼Ð¨ç¼ÆÈ¡A¦]¦¹feasible solution¬O¨D¥Xupper bound¡A¦Ó¥B§Ṳ́£Â_
¦a¹Á¸Õ°§Cupper bound)
C. ¦P¤W ÃD¡A¦ý¥Ø¼Ð§ï¬°¨D ¥X¥i¥H©ñ¤JI
¥]¤¤ª««~§Q¼íªº³Ì¤j¤Æ¡A¨Ï¥Î¤À¤ä©w¬Éºtºâªk¨Ó¸Ñ¦¹0/1I¥]°Ý ÃD¡C ¡]µù:
¥²¶·µe¥X·j´M¾ð¡^(¦¹ÃD¨Ï¥Î¥¿ªº¥Ø¼Ð¨ç¼ÆÈ¡A¦]¦¹feasible solution¬O¨D¥Xlower
bound¡A¦Ó¥B§Ṳ́£Â_¦a¹Á¸Õ´£°ªlower bound)
D.
¨Ï¥ÎA*ºtºâªk¨Ó¸Ñ¥H¤U¤§¦h¶¥¹Ï³Ìµu¸ô®|°ÝÃD¡C¡]µù: ¥²¶·µe¥X·j´M¾ð¡^
E. ±NAÃDªºS»PT¹ï½Õ¡A½bÀY¤Ï¦V¸ÑÃD¡C
F. ±NBÃD©Ò¦³¦¨¥»¼¥H2´î3¸ÑÃD¡C
G. ±NCÃDªºª««~«¶q§ï¬°20, 5, 6¸ÑÃD¡C
H. ±NDÃDªºV0»PV8¹ï½Õ¡A½bÀY¤Ï¦V¸Ñ¥ÑV8¨ìV0ªº³Ì¬q¸ô®|°ÝÃD¡C
¥H¤U§ï¬°¹êÅé¤W½Ò¡A¦ý¬O¤W½Ò¹êªp¨ÌµM¿ý¼v¤W¶Ç¡C
- 11. °Ý
ÃD¤U¬É»P°ÝÃD¤ÀÃþ: P¡BNP¡BNP§xÃø»PNP§¹¥þ°ÝÃD (6/2)
*¤@Ó°ÝÃD
ªº¤U¬É¬°¥ô¦ó¯à¸Ñ¨M¦¹°ÝÃDªººtºâªk¦Ü¤Ö©Ò»Ýªº®É¶¡½ÆÂø«×¡C(The
lower
bound
of a problem is the least time complexity required for any
algorithm which can be used to solve this problem.)
*NP§¹¥þ°ÝÃD²z½×:
Y¥ô¦ó¤@ÓNP§¹¥þ°ÝÃD¥i¦b¦h¶µ¦¡®É¶¡³Q¸Ñ¨M¡A«h¨C¤@ÓNP°ÝÃD¬Ò¥i¦b¦h¶µ¦¡®É¶¡Àò±o¸Ñ¨M(¤]´N¬OP=NP)¡C
If any one NP-complete problem can be solved in polynomial
time, then every problem in NP can also be solved in
polynomial time (i.e., P=NP). (ProblemLB.zip)(Recipe, Cook and Carp)(NPC.zip)
*(NPC²z½×´úÅç¤Î¸Ñµª)
Problem Set 11:
A. ¥H¤Uªº±Ôz¬O¹ïÁÙ¬O¿ù¡A¨Ã¸ÑÄÀ¹ï©Î¿ùªºì¦]:
Y§Ú̯àÃÒ©ú¥ô¦ó¤@Ó NPC°ÝÃDªº³ÌÃaª¬ªp°ÝÃD¤U¬É(worst case problem lower
bound)¬O«ü¼Æ¨ç¼Æ¶q¯Å¡A«h§Ṳ́w¸gÃÒ©ú NP¤£µ¥©óP¡C
B. ¥H¤Uªº±Ôz¬O¹ïÁÙ¬O¿ù¡A¨Ã¸ÑÄÀ¹ï©Î¿ùªºì¦]:
Y§Ú̯àÃÒ©ú¥ô¦ó¤@Ó NPC°ÝÃDªº³ÌÃaª¬ªp°ÝÃD¤U¬É(worst case problem lower
bound)¬O¦h¶µ¦¡¨ç¼Æ¶q¯Å¡A«h§Ṳ́w¸gÃÒ©ú NPµ¥©óP¡C
C. ¥H¤Uªº±Ôz¬O¹ïÁÙ¬O¿ù¡A¨Ã¸ÑÄÀ¹ï©Î¿ùªºì¦]:
Y§Ú̯ઽ§ä¨ì¤@Ó½T©w©Êºtºâªk¡A¦b³Ì®tª¬ªp¤U¥H¦h¶µ¦¡®É¶¡½ÆÂø«×¸Ñ¨M¤@ÓNPC°ÝÃD¡A«h§Ṳ́w¸gÃÒ©úNPµ¥©óP¡C
D. ¥H¤Uªº±Ôz¬O¹ïÁÙ¬O¿ù¡A¨Ã¸ÑÄÀ¹ï©Î¿ùªºì¦]:
¤H̤w¸gÃÒ©ú: ¨S¦³¥ô¦ó½T©wºtºâªk(deterministic algorithm)¥i¥H¦b³Ì®tª¬ªp(worst
case)¤U¡A¥H¦h¶µ¦¡®É¶¡½ÆÂø«×¸Ñ¨MNPC°ÝÃD¡C
E. ¥H¤Uªº±Ôz¬O¹ïÁÙ¬O¿ù¡A¨Ã¸ÑÄÀ¹ï©Î¿ùªºì¦]:
¤H̤w¸gÃÒ©ú: ¨S¦³¥ô¦ó½T©wºtºâªk(deterministic algorithm)¥i¥H¦b³Ì®tª¬ªp(worst
case)¤U¡A¥H¦h¶µ¦¡®É¶¡½ÆÂø«×¸Ñ¨MNP-hard°ÝÃD¡C
F. ¥H¤Uªº±Ôz¬O¹ïÁÙ¬O¿ù¡A¨Ã¸ÑÄÀ¹ï©Î¿ùªºì¦]:
¥ô¦óNPC°ÝÃD¥i¥Hpolynomially reduces to¥t¤@ÓNPC°ÝÃD¡C
G. ÃÒ©ú¤ä°t¶°°ÝÃD(dominating set problem)¬°NP°ÝÃD¡CÃÒ©ú¤ä°t¶°°ÝÃD(dominating set
problem)¬°NP°ÝÃD¡C
H. ÃÒ©úÂIÂл\°ÝÃD(vertex cover problem)¬°NP°ÝÃD¡C
I. ÃÒ©ú¶°¹Î°ÝÃD(clique problem)¬°NP°ÝÃD¡C
J. ÃÒ©úµÛ¦â°ÝÃD(chromatic coloring problem)¬°NP°ÝÃD¡C
K. ÃÒ©ú0/1 I¥]°ÝÃD(0/1 knapsack problem)¬°NP°ÝÃD¡C
L. ÃÒ©ú3-º¡¨¬°ÝÃD(3-SAT problem)¬°NP°ÝÃD¡C
M. ÃÒ©ú³Ì¤j³Î°ÝÃD(Max cut problem)¬°NP°ÝÃD¡C
N. ÃÒ©ú¥v©Z·S¾ð°ÝÃD(Steiner tree problem)¬°NP°ÝÃD¡C
O. ÃÒ©ú¹º¤À°ÝÃD(partition problem)¬°NP°ÝÃD¡C
P. ÃÒ©úÀ»¤¤¶°¦X°ÝÃD(hitting set problem)¬°NP°ÝÃD¡C
Q. ÃÒ©ú¤Tºû¤Ç°t°ÝÃD(3-dimensional matching problem)¬°NP°ÝÃD¡C
- 12.
³Ì¤p¦¨¥»³Ì¤j¬y¶qºtºâªk(Minimum-Cost Maximum-Flow
Alg., Min-Cost Max-Flow Alg., MCMF Alg.)(MCMF-New.zip)
(Optional)
À³¥Î ¹ê¨Ò 1: Paper: Yung-Liang Lai and
Jehn-Ruey Jiang, "Sink-Connected Barrier Coverage Optimization
for Wireless Sensor Networks," in Proc. of 2011 International
Conference on Wireless and Mobile Communications (ICWMC 2011),
2011. (Slides)
À³¥Î¹ê¨Ò 2: Paper: Yung-Liang Lai, and Jehn-Ruey Jiang, "Barrier
Coverage with Optimized Quality for Wireless Sensor Networks,"
in Proc. of the 15th International Symposium on Wireless
Personal Multimedia Communications Symposium (WPMC'12), 2012.
(Slides)
À³¥Î¹ê¨Ò 3: Jehn-Ruey Jiang,
Guan-Yi Sung, and Jih-Wei Wu, "LOM:
A Leader Oriented Matchmaking Algorithm for Multiplayer
Online Games," in Proc. of International
Conference on Internet Studies (NETs 2015), 2015.
(Distinguished Paper Award)(LOM.ppt)
*¦I¤ú§Qºtºâªk(Hungarian
Algorithm)(HungarianAlg.zip)(6/9)
Problem
Set
12: (Optional)
A.¥HFord-
Fulkersonºtºâªk (·f°tHill-Climbing
Search)¸Ñ¨M¥H¤U³Ì¤j¬y¶q°ÝÃD(¹Ïªº¦³¦VÃä(directed edge)¤W©Ò¼Ð¥Üªº¬°®e¶q(capacity))
B. ¥HEdmonds
-Karpºtºâªk¸Ñ¨M¥H¤U³Ì¤j¬y¶q°ÝÃD(¹Ïªº¦³¦VÃä(directed edge)¤W©Ò¼Ð¥Üªº¬°®e¶q(capacity))
C.
¦Û¦æ³]p¤@Ó¬yºô (flow
network)(°£s¡Bt¥~¨ã¦³4Ó¸`ÂI¡A¨Ã¨ã¦³10Óedge)¡A¥HEdmonds-Karpºtºâªk¸Ñ¨M
¨ä³Ì¤j¬y¶q°ÝÃD¡C
D.
§¹¦¨§ë¼v¤ùp30¤§Bellman-FordºtºâªkÀË´út¥[Åv´`Àô½d¨Ò¨ìiteration 8
E. §¹¦¨§ë¼v¤ù
p30¤§Bellman- FordºtºâªkÀË´út¥[Åv´`Àô½d¨Ò¨ìiteration 8¡A¨Ã¥[¤Jpredecessorµù°O
F. »¡©ú¦p¦ódetect§ë¼v¤ùp30¤§t¥[Åv´`Àô
Problem Set 12: (*A
and *B for 6/9)
*A. ¨Ï¥Î¦I¤ú§Qºtºâªk(Hungarian
algorithm)¨Ó¸Ñ¥H¤Uªº«ü¬£°ÝÃD(assignment problem)¡]µù:
¥²¶·¼g¥Xºtºâªk°õ¦æ¹Lµ{¤¤ªº¨CÓ¤¤¶¡µ²ªG¡^
|
Task A
|
Task B
|
Task C
|
Tim
|
$1
|
$2
|
$3
|
Bob
|
$2
|
$3
|
$2
|
Alex
|
$2
|
$2
|
$3
|
*B. ¦Û¦æ³]p¤@Ó«ü ¬£°ÝÃD(assignment
problem)ªº4x4¦¨¥»¯x°}¡A¨Ã¥H¦I¤ú§Qºtºâªk(Hungarian
algorithm)§ä¥X³Ì¤jÅv«§¹¬ü¤G¤À¤Ç°t(Maximum-Weight Perfect Bipartite
Matching)¡C
C. »¡©ú¦p¦ó¨Ï¥Î¦I¤ú§Qºtºâªk(Hungarian
algorithm)¸Ñ¨M³Ì¤p¼Ú¤ó¥±Åv«°t¹ï(Minimum Euclidean Weighted
Matching)°ÝÃD¡C©Ò¿×³Ì¤p¼Ú¤ó¥±Åv« °t¹ï°ÝÃD´yz¦p¤U:
µ¹©wnÓÂI(n¬°°¸¼Æ)¡A¦p¦ó±N¦¹nÓÂI¤Ç°t§Î¦¨n/2ÓÂI¹ï¡AÅý¨CÓÂI¹ï§Î¦¨¤@±ø½u¬q¡A¦Ó¦¹n/2±ø½u¬q¨ã¦³³Ì¤pªºªø«×Á`©M¡C
D. ÀH·Nµe¥X¥|Ó¼Ú¤ó¥±ÂI(2D
point)¡A³]¨ä¶ZÂ÷¬Ò¬°¾ã¼Æ(¥H¤½¤Àpºâ)¡A¥H¦I¤ú§Qºtºâªk(Hungarian
algorithm)§ä¥X¨ä³Ì¤p¼Ú¤ó¥±Åv«°t¹ï(Minimum Euclidean Weighted
Matching)¡C
E.
»¡©ú¦p¦ó±N¦I¤ú§Qºtºâªk¯à°÷¸Ñµªªº³Ì
¤p¦¨¥»«ü ¬£°ÝÃD(assignment problem)ÅÜÂର(reduce to)³Ì¤j¬y°ÝÃD
F. ±N¥H ¤U³Ì¤p¦¨¥»«ü ¬£°ÝÃD(assignment problem)Âà´«¬°¬yºô(flow
network)¡A¥H«K¨Ï¥Î³Ì¤j¬y(max-flow)ºtºâªk¸Ñ¨M¤§
|
Task 1
|
Task 2
|
Task 3
|
Carl
|
$24
|
$28
|
$24
|
Bob
|
$26
|
$32
|
$28
|
Alex
|
$24
|
$28
|
$30
|
- 13. (Optional) ªñ¦üºtºâªk
(Approximation algorithms) (ApproximationAlgorithm.zip)
¼Ú©Ô®È³~ºtºâªk (Eulerian Tour Algorithm) (EulerianTour.zip)
©w²z¤@
*³s³qªºµL¦V¹ÏG¦³¼Ú©Ô¸ô®|ªº¥Rn±ø¥ó¬O¡G G¤¤©_³»ÂI¡]³s±µªºÃä¼Æ¶q¬°©_¼Æªº³»ÂI¡^ªº¼Æ¥Øµ¥©ó0©ÎªÌ2¡C
*³s³qªºµL¦V¹ÏG¬O¼Ú©ÔÀô¡]¦s¦b¼Ú©Ô°j¸ô¡^ªº¥Rn±ø¥ó¬O¡G G¤¤¨CÓ³»ÂIªº«×³£¬O°¸¼Æ¡C
©w²z¤G
*¤@Ó³s³qªº¦³¦V¹Ï¥i¥Hªí¥Ü¬°¤@±ø±q³»ÂI u¨ì vªº¡]¤£³¬¦Xªº¡^¼Ú©Ô¸ô®|ªº¥Rn±ø¥ó¬O¡G
uªº¥X«×¡]±q³oÓ³»ÂIµo¥Xªº¦³¦VÃ䪺¼Æ¶q¡^¤ñ¤J«×¡]«ü¦V³oÓ³»ÂIªº¦³¦VÃ䪺¼Æ¶q¡^¦h1¡A
vªº¥X«×¤ñ¤J«×¤Ö1¡A¦Ó¨ä¥¦³»ÂIªº¥X«×©M¤J«×³£¬Ûµ¥¡C
*¤@Ó³s³qªº¦³¦V¹Ï¬O¼Ú©ÔÀô¡]¦s¦b¼Ú©Ô°j¸ô¡^ªº¥Rn±ø¥ó¬O¡G¨CÓ³»ÂIªº¥X«×©M¤J«×³£¬Ûµ¥¡C
Problem
Set
13:
A. ¨Ï¥Î2-ªñ¦üºtºâªk(2-approximation algorithm)¨Ó¸Ñ¨M¥H¤U¹Ïªº³»ÂIÂл\°ÝÃD(vertex cover problem)¡C¡]µù: ¶·´yz¦b°õ¦æºtºâªk¹Lµ{ªº¨CÓ¤¤¶¡µ²ªG¡C¡^
B. ¨Ï¥Î¼Ú©Ô®È³~ºtºâªk¨Ó§ä¥X¥H¤U¹Ïªº¼Ú©Ô®È³~¡C¡]µù: ¶·´yz¦b°õ¦æºtºâªk¹Lµ{ªº¨CÓ¤¤¶¡µ²ªG¡C¡^
C. ±N¥H¦ÛµM»y¨¥¼¶¼gªº¡uµL¦V¹Ï¼Ú©Ô®È³~/¸ô®|ºtºâªk¡v§ï¬°¥HµêÀÀ½X¼¶¼g
D. ¥H¦ÛµM»y¨¥¼¶¼g¡u¦³¦V¹Ï¼Ú©Ô®È³~/¸ô®|ºtºâªk¡v
E. ¥HµêÀÀ½X¼¶¼g¡u¦³¦V¹Ï¼Ú©Ô®È³~/¸ô®|ºtºâªk¡v
F. ¤ÀªR¨âÓºtºâªkªº®É¶¡½ÆÂø«×