CSE 101- Winter 18 Discussion Section Week 2

Transcription

1 S 101- Winter 18 iscussion Section Week 2

2 Topics Topological ordering Strongly connected components Binary search ntroduction to ivide and onquer algorithms

3 Topological ordering

4 Topological ordering => (only forward edges if vertices arranged in increasing order of labels) B

5 Topological ordering

6 Topological ordering inductive approach ow would you approach this problem?

7 Topological ordering inductive approach ow would you approach this problem? ind a vertex which you know can be labelled 1 What are the properties of such a starting vertex?

8 Topological ordering inductive approach ow would you approach this problem? ind a vertex which you know can be labelled 1 What are the properties of such a starting vertex? laim: always has some vertex with indegree = 0 Take an arbitrary vertex v. f v doesn t have indegree = 0, traverse any incoming edge to reach a predecessor of v. f this vertex doesn t have indegree = 0, traverse any incoming edge to reach a predecessor, etc. ventually, this process will either dentify a vertex with indegree = 0, or else reach a vertex that has been reached previously (a contradiction, given that is acyclic)

9 Topological ordering inductive approach ow would you approach this problem? ind a vertex which you know can be labelled 1 What are the properties of such a starting vertex? nductive (or recursive) approach ind a vertex v with indegree(v) = 0; give it lowest available label; elete v (and incident edges, update degrees of remaining edges) Repeat (Bonus) can you do the same, beginning with an ending vertex?

10 Topological ordering using S laim: n a, every edge leads to a vertex with lower post number. Why?

11 Topological ordering using S laim: n a, every edge leads to a vertex with lower post number. Why? ny edge (u,v) for which post(v) > post(u) is a back edge. But a, being acyclic, has no back edges.

12 Topological ordering using S laim: n a, every edge leads to a vertex with lower post number. Why? ny edge (u,v) for which post(v) > post(u) is a back edge. But a, being acyclic, has no back edges. Obvious solution: Run S, then perform tasks in decreasing order of post numbers Linear running time!

13 Topological ordering example B

14 Topological ordering example 1, B

15 Topological ordering example 1, B 2,3

16 Topological ordering example 1, 4,5 B 2,3

17 Topological ordering example 1,6 4,5 B 2,3

18 Topological ordering example 1,6 4,5 B 2,3 Vertices explored in alphabetical order Topological order: --B

19 Topological ordering example 1,6 4,5 B 2,3 Vertices explored in alphabetical order Topological order: --B B Vertices explored in order: B--

20 Topological ordering example 1,6 4,5 B 2,3 Vertices explored in alphabetical order Topological order: --B B 1, Vertices explored in order: B--

21 Topological ordering example 1,6 4,5 B 2,3 Vertices explored in alphabetical order Topological order: --B B 1,2 Vertices explored in order: B--

22 Topological ordering example 1,6 4,5 B 2,3 Vertices explored in alphabetical order Topological order: --B 3, B 1,2 Vertices explored in order: B--

23 Topological ordering example 1,6 4,5 B 2,3 Vertices explored in alphabetical order Topological order: --B 3,4 B 1,2 Vertices explored in order: B--

24 Topological ordering example 1,6 4,5 B 2,3 Vertices explored in alphabetical order Topological order: --B 5, 3,4 B 1,2 Vertices explored in order: B--

25 Topological ordering example 1,6 4,5 B 2,3 Vertices explored in alphabetical order Topological order: --B 5,6 3,4 B 1,2 Vertices explored in order: B-- Topological order: --B

26 Strongly onnected omponents Two vertices u and v of a directed graph are connected if there is a path from u to v and a path from v to u.

27 Strongly onnected omponents Two vertices u and v of a directed graph are connected if there is a path from u to v and a path from v to u. This definition leads to a partition of a directed graph into disjoint sets of vertices -> strongly connected components B

28 Meta graph Meta raph The graph obtained by shrinking each strongly connected component down to a single meta-node and drawing an edge from one meta-node to another if there is an edge between their respective components. Meta raph is a. (Why?) n other words, every directed graph is a dag of its strongly connected components. B

29 Meta graph Meta raph The graph obtained by shrinking each strongly connected component down to a single meta-node and drawing an edge from one meta-node to another if there is an edge between their respective components. Meta raph is a. (Why?) n other words, every directed graph is a dag of its strongly connected components. B

30 Meta graph Meta raph The graph obtained by shrinking each strongly connected component down to a single meta-node and drawing an edge from one meta-node to another if there is an edge between their respective components. Meta raph is a. (Why?) n other words, every directed graph is a dag of its strongly connected components., B,,, B

31 Binary search Looking up a word in the dictionary Problem: ind x in a sorted array [1 n] lgorithm: ompare x with middle element of array Recursively find x in left subarray or right subarray

32 Binary search pseudocode Binary_search([], key, first, last) if (first > last) return not_found mid = (first + last)/2 if (key == [mid]) return mid if (key < [mid]) return Binary_search([],key,first,mid-1) if (key > [mid]) return Binary_search([],key,mid+1,last)

33 ivide and onquer ivide the problem (instance) into one or more subproblems onquer each subproblem recursively ombine separate solutions

34 Binary search pseudocode Binary_search([], key, first, last) if (first > last) return not_found //base case mid = (first + last)/2 if (key == [mid]) return mid //divide & conquer if (key < [mid]) return Binary_search([],key,first,mid-1) if (key > [mid]) return Binary_search([],key,mid+1,last)

35 Binary search revisited ivide: compare x with middle element onquer: recurse in one subarray ombine: trivial Running time T(n) = T(n/2) + O(1) Master theorem: T(n) a T(n/b) + O(n d ) 1. T(n) = O(n d ) if a < b d 2. T(n) = O(n d log n) if a = b d 3. T(n) = O(nlogba) if a > b d Using master theorem, T(n) = O(log n) (ase 2)

36 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. a) ndicate the pre and post numbers of the nodes. Solution

37 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. a) ndicate the pre and post numbers of the nodes. Solution 1,

38 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. a) ndicate the pre and post numbers of the nodes. Solution 1, 2,

39 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. a) ndicate the pre and post numbers of the nodes. Solution 1, 3, 2,

40 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. a) ndicate the pre and post numbers of the nodes. Solution 1, 2, 3, 4,

41 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. a) ndicate the pre and post numbers of the nodes. Solution 1, 2, 3, 4, 5,

42 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. a) ndicate the pre and post numbers of the nodes. Solution 1, 2, 3, 4, 5,6

43 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. a) ndicate the pre and post numbers of the nodes. Solution 1, 2, 3, 4, 5,6 7,

44 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. a) ndicate the pre and post numbers of the nodes. Solution 1, 2, 3, 4, 5,6 7,8

45 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. a) ndicate the pre and post numbers of the nodes. Solution 1, 2, 3, 4,9 5,6 7,8

46 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. a) ndicate the pre and post numbers of the nodes. Solution 1, 2, 3,10 4,9 5,6 7,8

47 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. a) ndicate the pre and post numbers of the nodes. Solution 1, 2, 3,10 11, 4,9 5,6 7,8

48 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. a) ndicate the pre and post numbers of the nodes. Solution 1, 2, 3,10 11,12 4,9 5,6 7,8

49 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. a) ndicate the pre and post numbers of the nodes. Solution 1, 2,13 3,10 11,12 4,9 5,6 7,8

50 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. a) ndicate the pre and post numbers of the nodes. Solution 1, 2,13 3,10 11,12 4,9 5,6 7,8

51 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. a) ndicate the pre and post numbers of the nodes. Solution 1, 14, 2,13 3,10 11,12 4,9 5,6 7,8

52 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. a) ndicate the pre and post numbers of the nodes. Solution 1, 14,15 2,13 3,10 11,12 4,9 5,6 7,8

53 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. B a) ndicate the pre and post numbers of the nodes. Solution 1,16 14,15 2,13 3,10 11,12 4,9 5,6 7,8

54 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. b) What are the sources and sinks of the graph?

55 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. b) What are the sources and sinks of the graph? Solution rom the S performed in the previous step we are aware that vertex is a source node since it has the highest post number.

56 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. b) What are the sources and sinks of the graph? Solution rom the S performed in the previous step, we are aware that vertex has the lowest post number. Thus, vertex is a sink.

57 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. b) What are the sources and sinks of the graph? Solution rom the S performed in the previous step, we are aware that vertex has the lowest post number. Thus, vertex is a sink. lso, in the above S while visiting the vertices in the adjacency list of vertex we could have visited vertex before vertex. This would have lead to vertex having the least post number. Thus, vertex can be a sink as well.

58 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. c) What topological ordering is found by the algorithm? Solution

59 Solved xercise Run the S-based topological ordering algorithm on the following graph. Whenever you have a choice of vertices to explore, always pick the one that is alphabetically first. c) What topological ordering is found by the algorithm? Solution