Path with Maximum Probability
描述
You are given an undirected weighted graph of n
nodes (0-indexed), represented by an edge list where edges[i] = [a, b]
is an undirected edge connecting the nodes a and b with a probability of success of traversing that edge succProb[i]
.
Given two nodes start
and end
, find the path with the maximum probability of success to go from start
to end
and return its success probability.
If there is no path from start
to end
, return 0. Your answer will be accepted if it differs from the correct answer by at most 1e-5.
Example 1:
Input: n = 3, edges = [[0,1],[1,2],[0,2]], succProb = [0.5,0.5,0.2], start = 0, end = 2
Output: 0.25000
Explanation: There are two paths from start to end, one having a probability of success = 0.2 and the other has 0.5 * 0.5 = 0.25.
Example 2:
Input: n = 3, edges = [[0,1],[1,2],[0,2]], succProb = [0.5,0.5,0.3], start = 0, end = 2
Output: 0.30000
Example 3:
Input: n = 3, edges = [[0,1]], succProb = [0.5], start = 0, end = 2
Output: 0.00000
Explanation: There is no path between 0 and 2.
Constraints:
- 0 <= start, end < n
- start != end
- 0 <= a, b < n
- a != b
- 0 <= succProb.length == edges.length <=
- 0 <= succProb[i] <= 1
- There is at most one edge between every two nodes.
分析
由于每条边的概率是介于[0,1]之间的正数,且路径上的概率是累乘起来的,那么对原图G
中的每条边的权重取对数然后取反,就得到了一个边权在之间的图G'
,图G
中“从起点到终点成功概率最大”的路径对应了图G'
中“从起点到终点边权之和最小”的路径。由于图G'
中没有负数边权,用 Dijkstra 算法最合适不过了。
代码
- Java
- C++
// Path with Maximum Probability
// Dijkstra
// Time Complexity: O(ElogN), Space Complexity: O(N + E)
class Solution {
public double maxProbability(int n, int[][] edges, double[] succProb, int start, int end) {
// adjacency list, map<vertex_id, map<vertex_id, weight>>
Map<Integer, Map<Integer, Double>> graph = new HashMap<>();
for (int i = 0; i < edges.length; i++) {
int[] edge = edges[i];
double w = -Math.log(succProb[i]);
// Undirected
graph.putIfAbsent(edge[0], new HashMap<>());
graph.get(edge[0]).put(edge[1], w);
graph.putIfAbsent(edge[1], new HashMap<>());
graph.get(edge[1]).put(edge[0], w);
}
Map<Integer, Double> dist = dijkstra(graph, start);
return dist.containsKey(end) ? Math.exp(-dist.get(end)) : 0;
}
/** Standard Dijkstra algorithm.
*
@param graph Adjacency list, map<vertex_id, map<vertex_id, weight>>.
@param start The starting vertex ID.
@return dist, map<vertex_id, distance>.
*/
private static Map<Integer, Double> dijkstra(Map<Integer, Map<Integer, Double>> graph, int start) {
// map<vertex_id, distance>
Map<Integer, Double> dist = new HashMap<>();
// vertex_id -> father_vertex_id
Map<Integer, Integer> father = new HashMap<>();
// pair<distance, vertex_id>, min heap, sorted by distance from start to vertex_id
Queue<Pair<Double, Integer>> pq = new PriorityQueue<>((a, b) -> Double.compare(a.getKey(), b.getKey()));
// from start to start itself
pq.offer(new Pair(0.0, start));
dist.put(start, 0.0);
while(!pq.isEmpty()){
final int u = pq.poll().getValue();
if (!graph.containsKey(u)) continue; // leaf node
for(int v : graph.get(u).keySet()){
final double w = graph.get(u).get(v);
if (!dist.containsKey(v) || dist.get(u)+ w < dist.get(v)) {
final double shorter = dist.get(u)+ w;
dist.put(v, shorter);
father.put(v, u);
pq.offer(new Pair(shorter, v));
}
}
}
return dist;
}
}
// TODO