[ Algorithm ] 08. Minimum Spanning Trees

21900772·2023년 5월 20일

MST Minimum Spanning Trees algorithm 알고리즘

알고리즘 분석

목록 보기

8/14

🖥️ MST

Spanning Tree : Spanning graph + Tree(acycle)

Problem: given a connected, undirected, weighted graph, find a spanning tree using edges that minimize the total weight
Undirected graph G = (V, E)
Weight w(u,v) on each edge (u,v) $\in$ E
Spanning tree of G’ is a minimal subgraph of G such that
- V(G’) = V(G) and G’ is connected
- Any connected graph with n vertices must have at least n-1 edges. All connected graphs with n-1 edges are trees.
  ➡️ connected + n vertices + n-1 edges : cycle이 없음을 의미
Find T $\subseteq$ E s.t.
- T connects all vertices (T is a spanning tree), and
- w(T) = $\sum_{(u,v) \in E}$ w(u,v)(모든 edge weight의 합) is minimized
Has n -1 edges, no cycle, might not be unique
Example)
Interconnect n pins with n-1 wires, each connecting two pins so that we use the least amount of wire.
We’ll look at two greedy algorithms
- Kruskal’s algorithm
- Prim’s algorithm

Building up the solution

Build a set A of edges
Initially, A is empty
As we add edges to A(= n-1, exit), maintain a loop invariant : A is a subset of some MST
Add only edges that maintain the invariant(불변성)
- If A is a subset of some MST, an edge (u,v) is safe for A if and only if A U { (u,v) } is also a subset of some MST
- So, we will add only safe edges

Generic MST algorithm➡️ safe : looks best (smallest or largest)

Finding a safe edge

Let S $\subset$ V and A $\subseteq$ E
- A cut (S, V – S) is a partition of vertices into disjoint sets S and V – S
- Edge (u, v) $\in$ E crosses cut (S, V – S ) if one endpoint is in S and the other is in V – S
- A cut respects A if and only if no edge in A crosses the cut
  ➡️ cut을 지나가지 않음
- An edge is a light edge crossing a cut if and only if its weight is minimum over all edges crossing the cut. For a given cut, there can be more than one light edge crossing it (값이 같으면 여러개)

Theorem

Let A be a subset of E in some MST, (S, V – S) be a cut that respects A, and (u, v) be a light edge crossing (S, V – S )
Then, (u, v) is safe for A

Proof)
1. Let T be a MST that includes A, and assume that T does not contain the light edge (u, v)
2. We shall construct another MST T ́ that includes A U { (u,v) } .
3. Edge (u,v) forms a cycle with the edges on the path p from u to v in T
➡️ n -1 edge에서 n개가 되기 떄문
4. Since u and v are on opposite sides of the cut (S, V – S) there is at least one edge in T on the path p that also crosses the cut. Let (x,y) be any such edge.
5. Removing (x,y) breaks T into two components. And adding (u,v) reconnects them to form a new spanning tree T ́
w(T ́) = w(T) – w(x,y) + w(u,v) ≤ w(T)
Thus, T ́ must be a MST
And since A U { (u,v) } $\subseteq$ T ́, (u,v) is safe for A

Corollary

Let A be a subset of E that is included in some minimum spanning tree for G, and let C = (V $_c$ , E $_c$ ) be a connected component (tree) in the forest G $_A$ = (V, A). If (u,v) is a light edge connecting C to some other component in G $_A$ , then (u,v) is safe for A

Proof) The cut (Vc, V – Vc) respects A, and (u,v) is a light edge for this cut. Therefore, (u,v) is safe for A.

🖥️ Kruskal’s algorithm

Sort edges into nondecreasing(increasing) order by w.
The algorithm maintains A, a forest of trees ➡️ 마지막에 big one tree
Repeatedly merges two components into one by choosing the light edge that connects them
i.e.,
1. Choose the light edge crossing the cut between them.
2. (If it forms a cycle, the edge is discarded)
What is the design strategy of Kruskal’s algorithm?
- Greedy

pseudo-code

Example

🖥️ Prim's algorithm

Builds one tree, so A is always a tree (cf. kruskal is forest)
Start from an arbitrary “root” r
At each step, find a light edge crossing cut (V $_A$ , V – V $_A$ ), where V $_A$ = vertices that A is incident on
$\pi$ [v] = parent of v, NIL if it has no parent or v = r.
To find a light edge quickly
- use a priority queue Q