南开大学算法导论第三章
算法导论第二章答案
第二章算法入门 由于时间问题有些问题没有写的很仔细，而且估计这里会存在不少不恰当之处。另，思考题2-3 关于霍纳规则，有些部分没有完成，故没把解答写上去，我对其 c 问题有疑问，请有解答方法者提供个意见。 给出的代码目前也仅仅为解决问题，没有做优化，请见谅，等有时间了我再好好修改。 插入排序算法伪代码 INSERTION-SORT(A) 1 for j ← 2 to length[A] 2 do key ←A[j] 3 Insert A[j] into the sorted sequence A[1..j-1] 4 i ←j-1 5 while i > 0 and A[i] > key 6 do A[i+1]←A[i] 7 i ←i ? 1 8 A[i+1]←key C#对揑入排序算法的实现: public static void InsertionSort
算法导论 第三版 第21章 答案 英
Chapter21 Michelle Bodnar,Andrew Lohr April12,2016 Exercise21.1-1 EdgeP rocessed initial{a}{b}{c}{d}{e}{f}{g}{h}{i}{j}{k} (d,i){a}{b}{c}{d,i}{e}{f}{g}{h}{j}{k} (f,k){a}{b}{c}{d,i}{e}{f,k}{g}{h}{j} (g,i){a}{b}{c}{d,i,g}{e}{f,k}{h}{j} (b,g){a}{b,d,i,g}{c}{e}{f,k}{h}{j} (a,h){a,h}{b,d,i,g}{c}{e}{f,k}{j} (i,j){a,h}{b,d,i,g,j}{c}{e}{f,k} (d,k){a,h}{b,d,i,g,j,f,k}{c}{e} (b,j){a,h}{b,d,i,g,j,f,k}{c}{e} (d,f){a,h}{b,d,i,g,j,f,k}{c}{e} (g,j){a,h}{b,d,i,g,j,f,k}{c}{e} (a,e){a,h,e}{b,d,i,g,j,f,k}{c} So,the connected that we are left with are{a,h,e},{b,d,i,g,j,f,k}, and{c}. Exercise21.1-2 First suppose that two vertices are in the same connected component.Then there exists a path of edges connecting them.If two vertices are connected by a single edge,then they are put into the same set when that edge is processed. At some point during the algorithm every edge of the path will be processed,so all vertices on the path will be in the same set,including the endpoints.Now suppose two vertices u and v wind up in the same set.Since every vertex starts o?in its own set,some sequence of edges in G must have resulted in eventually combining the sets containing u and v.From among these,there must be a path of edges from u to v,implying that u and v are in the same connected component. Exercise21.1-3 Find set is called twice on line4,this is run once per edge in the graph,so, we have that?nd set is run2|E|times.Since we start with|V|sets,at the end 1
算法导论 第三版 第六章 答案 英
Chapter6 Michelle Bodnar,Andrew Lohr December31,2016 Exercise6.1-1 At least2h and at most2h+1?1.Can be seen because a complete binary tree of depth h?1hasΣh?1 i=02i=2h?1elements,and the number of elements in a heap of depth h is between the number for a complete binary tree of depth h?1exclusive and the number in a complete binary tree of depth h inclusive. Exercise6.1-2 Write n=2m?1+k where m is as large as possible.Then the heap consists of a complete binary tree of height m?1,along with k additional leaves along the bottom.The height of the root is the length of the longest simple path to one of these k leaves,which must have length m.It is clear from the way we de?ned m that m= lg n . Exercise6.1-3 If there largest element in the subtee were somewhere other than the root, it has a parent that is in the subtree.So,it is larger than it’s parent,so,the heap property is violated at the parent of the maximum element in the subtree Exercise6.1-4 The smallest element must be a a leaf node.Suppose that node x contains the smallest element and x is not a leaf.Let y denote a child node of x.By the max-heap property,the value of x is greater than or equal to the value of y.Since the elements of the heap are distinct,the inequality is strict.This contradicts the assumption that x contains the smallest element in the heap. Exercise6.1-5 Yes,it is.The index of a child is always greater than the index of the parent, so the heap property is satis?ed at each vertex. 1
算法导论 第三版 第七章 答案 英
Chapter7 Michelle Bodnar,Andrew Lohr April12,2016 Exercise7.1-1 13199512874212611 13199512874212611 13199512874212611 91913512874212611 95131912874212611 95131912874212611 95819121374212611 95871213194212611 95874131912212611 95874131912212611 95874219122113611 95874261221131911 95874261121131912 Exercise7.1-2 If all elements in the array have the same value,PARTITION returns r.To make PARTITION return q= (p+r)/2 when all elements have the same value,modify line4of the algorithm to say this:if A[j]≤x and j(mod2)= (p+1)(mod2).This causes the algorithm to treat half of the instances of the same value to count as less than,and the other half to count as greater than. Exercise7.1-3 The for loop makes exactly r?p iterations,each of which takes at most constant time.The part outside the for loop takes at most constant time.Since r?p is the size of the subarray,PARTITION takes at most time proportional to the size of the subarray it is called on. Exercise7.1-4 To modify QUICKSORT to run in non-increasing order we need only modify line4of PARTITION,changing≤to≥. 1
算法导论 第三版 第二章 答案 英
Chapter2 Michelle Bodnar,Andrew Lohr April12,2016 Exercise2.1-1 314159264158 314159264158 314159264158 263141594158 263141415958 263141415859 Exercise2.1-2 Algorithm1Non-increasing Insertion-Sort(A) 1:for j=2to A.length do 2:key=A[j] 3://Insert A[j]into the sorted sequence A[1..j?1]. 4:i=j?1 5:while i>0and A[i] Chapter24 Michelle Bodnar,Andrew Lohr April12,2016 Exercise24.1-1 If we change our source to z and use the same ordering of edges to decide what to relax,the d values after successive iterations of relaxation are: s t x y z ∞∞∞∞0 2∞7∞0 25790 25690 24690 Theπvalues are: s t x y z NIL NIL NIL NIL NIL z NIL z NIL NIL z x z s NIL z x y s NIL z x y s NIL Now,if we change the weight of edge(z,x)to4and rerun with s as the source,we have that the d values after successive iterations of relaxation are: s t x y z 0∞∞∞∞ 06∞7∞ 06472 02472 0247?2 Theπvalues are: s t x y z NIL NIL NIL NIL NIL NIL s NIL s NIL NIL s y s t NIL x y s t NIL x y s t 1 Note that these values are exactly the same as in the worked example.The di?erence that changing this edge will cause is that there is now a negative weight cycle,which will be detected when it considers the edge(z,x)in the for loop on line5.Since x.d=4>?2+4=z.d+w(z,x),it will return false on line7. Exercise24.1-2 Suppose there is a path from s to v.Then there must be a shortest such path of lengthδ(s,v).It must have?nite length since it contains at most|V|?1 edges and each edge has?nite length.By Lemma24.2,v.d=δ(s,v)<∞upon termination.On the other hand,suppose v.d<∞when BELLMAN-FORD ter-minates.Recall that v.d is monotonically decreasing throughout the algorithm, and RELAX will update v.d only if u.d+w(u,v) Chapter19 Michelle Bodnar,Andrew Lohr April12,2016 Exercise19.2-1 First,we take the subtrees rooted at24,17,and23and add them to the root list.Then,we set H.min to18.Then,we run consolidate.First this has its degree2set to the subtree rooted at18.Then the degree1is the subtree rooted at38.Then,we get a repeated subtree of degree2when we consider the one rooted at24.So,we make it a subheap by placing the24node under18. Then,we consider the heap rooted at17.This is a repeat for heaps of degree1, so we place the heap rooted https://www.wendangku.net/doc/9f10672282.html,stly we consider the heap rooted at23,and then we have that all the di?erent heaps have distinct degrees and are done,setting H.min to the smallest,that is,the one rooted at17. The three heaps that we end up with in our root list are: 23 17 38 30 41 and 1 算法设计与分析 学 习 报 告 第一部分学习内容归纳 “计算机算法是以一步接一步的方式来详细描述计算机如何将输入转化为所要求的输出的过程，或者说，算法是对计算机上执行的计算过程的具体描述。”（参考文献：百度百科）《算法设计与分析》是一门面向设计，在计算机科学中处于核心地位的课程。这门课程主要讲授了在计算机应用中经常遇到的问题和求解的方法，分治法、动态规划法、随机算法等设计算法的基本原理、技巧和算法复杂性的分析，以及计算理论简介。 第一部分“概论和数学准备”在简单了解了算法的基本概念和复杂性、研究步骤等几个重要知识点后，着重学习了算法的数学基础，包括生成函数、差方方程的求解等，主要适用于求解算法的时间复杂性。 “任何可以用计算机求解的问题所需要的计算时间都与其规模有关：问题的规模越小，解题所需的计算时间往往也越短，从而也就比较容易处理。”（参考文献：《计算机算法设计与分析（第3版）》）而第二部分介绍的算法常用技术之首——分治法就运用了这样的思想。分治法的要领在于Divide（子问题的划分）-Conquer（子问题的求解）-Combine（子问题解的组合）。由于子问题和原问题是同类的，递归的思想在分治法中显得尤其重要，它们经常同时运用在算法设计中。这部分内容从Select（求第k小元）算法，寻找最近点对算法和快速傅立叶变换FFT等实际应用中深化对分治法思想的理解，同时也强调了平衡思想的重要性。 第三部分“动态规划”与分治法类似，同样是把问题层层分解成规模越来越小的同类型的子问题。但与分治法不同的是，分治法中的子问题通常是相互独立的，而动态规划法中的子问题很多都是重复的，因此通常采用递推的方法以避免重复计算。然而，也不是所有的情况下都采用递推法，当有大量的子问题无需求解时，更好的方式是采用动态规划法的变形——备忘录方法。通常需要用到动态规划法求解的问题都具有子问题的高度重复性和最优子结构性质两大特征，这也是我们分析问题和设计算法时的关键点。最长公共子序列LCS问题和最优二分搜索树就是从动态规划法的两个主要特征角度分析问题，进而设计出相应的解决算法的。而这部分内容中的另一个问题——流水作业调度，则告诉我们采用动态规划时偶尔也得不到高效的算法，我们要学会将已有的知识灵活运用，适当加工。 第四部分“集合算法”中首先介绍了一种分析算法复杂度的手法——平摊分析（Amortized Analysis）。与之前我们所接触的算法分析方法即逐一考虑执行每条指令所需的时间复杂度再进行累加的方法不同，平摊分析是对若干条指令从整体角度考虑其时间复杂度，通过这样的方法获得的时间复杂度更加贴近实际的情况。平摊分析的主要方法有聚集方法，会计方法和势能方法。聚集方法将指令的时间复杂度分类计算再相加；会计方法采用了耗费提前计算的思想；势能方法引入了势函数的概念，从每步操作的数据结构状态和势函数的关系角度分析得出操作的平摊代价。“集合算法”这一部分主要分析了Union（合并集合）和Find （给出元素所在集合名）这两种运算。从上学期的《数据结构》课程的学习中，我们就已经发现集合和树之间的关系是密不可分的，我们经常用树结构来表示集合。而2-3树是一种特殊的每个内结点都只有2个或3个儿子的树，广泛的应用于可实现Member（查找）、Insert（插入）、Delete（删除）操作的数据结构——字典，可实现Insert、Delete、Union和Min（查找最小叶结点）的数据结构——可并堆，可实现Insert、Delete、Find、Concatenate（保序合并）和Split Chapter10 Michelle Bodnar,Andrew Lohr April12,2016 Exercise10.1-1 4 41 413 41 418 41 Exercise10.1-2 We will call the stacks T and R.Initially,set T.top=0and R.top=n+1. Essentially,stack T uses the?rst part of the array and stack R uses the last part of the array.In stack T,the top is the rightmost element of T.In stack R, the top is the leftmost element of R. Algorithm1PUSH(S,x) 1:if S==T then 2:if T.top+1==R.top then 3:error“over?ow” 4:else 5:T.top=T.top+1 6:T[T.top]=x 7:end if 8:end if 9:if S==R then 10:if R.top?1==T.top then 11:error“over?ow” 12:else 13:R.top=R.top?1 14:T[T.top]=x 15:end if 16:end if 1 Algorithm2POP(S) if S==T then if T.top==0then error“under?ow” else T.top=T.top?1. return T[T.top+1] end if end if if S==R then if R.top==n+1then error“under?ow” else R.top=R.top+1. return R[R.top?1] end if end if Exercise10.1-3 4 41 413 13 138 38 Exercise10.1-4 Algorithm3ENQUEUE if Q.head==Q.tail+1,or Q.head==1and Q.tail==Q.length then error“over?ow” end if Q[Q.tail]=x if Q.tail==Q.length then Q.tail=1 else Q.tail=Q.head+1 end if Exercise10.1-5 As in the example code given in the section,we will neglect to check for over?ow and under?ow errors. 2 Chapter2 Getting Start 2.1 Insertion sort 2.1.2 将Insertion-Sort 重写为按非递减顺序排序 2.1.3 计算两个n 位的二进制数组之和 2.2 Analyzing algorithms 当前n-1个元素排好序后，第n 个元素已经是最大的元素了. 最好时间和最坏时间均为2()n Θ 2.3 Designing algorithms 2.3.3 计算递归方程的解 22()2(/2)2,1k if n T n T n n if n for k =?=?+ = >? （1） 当1k =时，2n =，显然有()lg T n n n = （2） 假设当k i =时公式成立，即()lg 2lg 22i i i T n n n i ===?， 则当1k i =+，即12i n +=时， 2.3.4 给出insertion sort 的递归版本的递归式 2.3-6 使用二分查找来替代insertion-sort 中while 循环内的线性扫描，是否可以将算法的时间提高到(lg )n n Θ？ 虽然用二分查找法可以将查找正确位置的时间复杂度降下来，但 是移位操作的复杂度并没有减少，所以最坏情况下该算法的时间复杂度依然是2()n Θ 2.3-7 给出一个算法，使得其能在(lg )n n Θ的时间内找出在一个n 元素的整数数组内，是否存在两个元素之和为x 首先利用快速排序将数组排序，时间(lg )n n Θ，然后再进行查找： Search(A,n,x) QuickSort(A,n); i←1; j←n; while A[i]+A[j]≠x and i Chapter16 Michelle Bodnar,Andrew Lohr April12,2016 Exercise16.1-1 The given algorithm would just stupidly compute the minimum of the O(n) numbers or return zero depending on the size of S ij.There are a possible number of subproblems that is O(n2)since we are selecting i and j so that 1≤i≤j≤n.So,the runtime would be O(n3). Exercise16.1-2 This becomes exactly the same as the original problem if we imagine time running in reverse,so it produces an optimal solution for essentially the same reasons.It is greedy because we make the best looking choice at each step. Exercise16.1-3 As a counterexample to the optimality of greedily selecting the shortest, suppose our activity times are{(1,9),(8,11),(10,20)}then,picking the shortest ?rst,we have to eliminate the other two,where if we picked the other two instead,we would have two tasks not one. As a counterexample to the optimality of greedily selecting the task that con?icts with the fewest remaining activities,suppose the activity times are {(?1,1),(2,5),(0,3),(0,3),(0,3),(4,7),(6,9),(8,11),(8,11),(8,11),(10,12)}.Then, by this greedy strategy,we would?rst pick(4,7)since it only has a two con- ?icts.However,doing so would mean that we would not be able to pick the only optimal solution of(?1,1),(2,5),(6,9),(10,12). As a counterexample to the optimality of greedily selecting the earliest start times,suppose our activity times are{(1,10),(2,3),(4,5)}.If we pick the ear-liest start time,we will only have a single activity,(1,10),whereas the optimal solution would be to pick the two other activities. Exercise16.1-4 Maintain a set of free(but already used)lecture halls F and currently busy lecture halls B.Sort the classes by start time.For each new start time which you encounter,remove a lecture hall from F,schedule the class in that room, 1 Chapter17 Michelle Bodnar,Andrew Lohr April12,2016 Exercise17.1-1 It woudn’t because we could make an arbitrary sequence of MULT IP USH(k),MULT IP OP(k). The cost of each will beΘ(k),so the average runtime of each will beΘ(k)not O(1). Exercise17.1-2 Suppose the input is a1followed by k?1zeros.If we call DECREMENT we must change k entries.If we then call INCREMENT on this it reverses these k changes.Thus,by calling them alternately n times,the total time isΘ(nk). Exercise17.1-3 Note that this setup is similar to the dynamic tables discussed in section 17.4.Let n be arbitrary,and have the cost of operation i be c(i).Then, n i=1c(i)= lg(n) i=1 2i+ i≤n not a power of2 1≤ lg(n) i=1 2i+n=21+ lg(n) ?1+n≤4n?1+n≤5n∈O(n) So,since to?nd the average,we divide by n,the average runtime of each com-mand is O(1). Exercise17.2-1 To every stack operation,we charge twice.First we charge the actual cost of the stack operation.Second we charge the cost of copying an element later on.Since we have the size of the stack never exceed k,and there are always k operations between backups,we always overpay by at least enough.So,the ammortized cost of the operation is constant.So,the cost of the n operation is O(n). Exercise17.2-2 1 Chapter35 Michelle Bodnar,Andrew Lohr April12,2016 Exercise35.1-1 We could select the graph that consists of only two vertices and a single edge between them.Then,the approximation algorithm will always select both of the vertices,whereas the minimum vertex cover is only one vertex.more generally,we could pick our graph to be k edges on a graph with2k vertices so that each connected component only has a single edge.In these examples,we will have that the approximate solution is o?by a factor of two from the exact one. Exercise35.1-2 It is clear that the edges picked in line4form a matching,since we can only pick edges from E ,and the edges in E are precisely those which don’t share an endpoint with any vertex already in C,and hence with any already-picked edge. Moreover,this matching is maximal because the only edges we don’t include are the ones we removed from E .We did this because they shared an endpoint with an edge we already picked,so if we added it to the matching it would no longer be a matching. Exercise35.1-3 We will construct a bipartite graph with V=R∪L.We will try to construct it so that R is uniform,not that R is a vertex cover.However,we will make it so that the heuristic that the professor(professor who?)suggests will cause us to select all the vertices in L,and show that|L|>2|R|. Initially start o?with|R|=n?xed,and L empty.Then,for each i from 2up to n,we do the following.Let k= n i .Select S a subset of the vertices of R of size ki,and so that all the vertices in R?S have a greater or equal degree.Then,we will add k vertices to L,each of degree i,so that the union of their neighborhoods is S.Note that throughout this process,the furthest apart the degrees of the vertices in R can be is1,because each time we are picking the smallest degree vertices and increasing their degrees by1.So,once this has been done for i=n,we can pick a subset of R whose degree is one less than the rest of R(or all of R if the degrees are all equal),and for each vertex in 1算法导论 第三版 第24章 答案 英
算法导论 第三版 第十九章 答案 英
算法导论学习报告
Ch10算法导论 第三版 第十章 答案 英
算法导论习题答案
算法导论 第三版 第十六章 答案 英
算法导论 第三版 第十七章 答案 英
算法导论 第三版 第35章 答案 英