Uniform-Cost Search es una variante del algoritmo de Dijikstra . Aquí, en lugar de insertar todos los vértices en una cola de prioridad, insertamos solo la fuente, luego insertamos uno por uno cuando sea necesario. En cada paso, verificamos si el elemento ya está en la cola de prioridad (usando la array visitada). En caso afirmativo, ejecutamos la tecla de disminución, de lo contrario, la insertamos.
Esta variante de Dijkstra es útil para gráficos infinitos y aquellos gráficos que son demasiado grandes para representarlos en la memoria. La búsqueda de costo uniforme se utiliza principalmente en inteligencia artificial.
Ejemplos:
Input :
Output : Minimum cost from S to G is =3
La búsqueda de costo uniforme es similar al algoritmo de Dijikstra. En este algoritmo desde el estado inicial, visitaremos los estados adyacentes y elegiremos el estado menos costoso, luego elegiremos el siguiente estado menos costoso de todos los estados no visitados y adyacentes de los estados visitados, de esta manera intentaremos alcanzar el estado objetivo (tenga en cuenta que no continuaremos el camino a través de un estado objetivo), incluso si alcanzamos el estado objetivo, continuaremos buscando otros caminos posibles (si hay múltiples objetivos). Mantendremos una cola de prioridad que dará el siguiente estado menos costoso de todos los estados adyacentes de los estados visitados.
C++
// C++ implementation of above approach #include <bits/stdc++.h> using namespace std; // graph vector<vector<int> > graph; // map to store cost of edges map<pair<int, int>, int> cost; // returns the minimum cost in a vector( if // there are multiple goal states) vector<int> uniform_cost_search(vector<int> goal, int start) { // minimum cost upto // goal state from starting // state vector<int> answer; // create a priority queue priority_queue<pair<int, int> > queue; // set the answer vector to max value for (int i = 0; i < goal.size(); i++) answer.push_back(INT_MAX); // insert the starting index queue.push(make_pair(0, start)); // map to store visited node map<int, int> visited; // count int count = 0; // while the queue is not empty while (queue.size() > 0) { // get the top element of the // priority queue pair<int, int> p = queue.top(); // pop the element queue.pop(); // get the original value p.first *= -1; // check if the element is part of // the goal list if (find(goal.begin(), goal.end(), p.second) != goal.end()) { // get the position int index = find(goal.begin(), goal.end(), p.second) - goal.begin(); // if a new goal is reached if (answer[index] == INT_MAX) count++; // if the cost is less if (answer[index] > p.first) answer[index] = p.first; // pop the element queue.pop(); // if all goals are reached if (count == goal.size()) return answer; } // check for the non visited nodes // which are adjacent to present node if (visited[p.second] == 0) for (int i = 0; i < graph[p.second].size(); i++) { // value is multiplied by -1 so that // least priority is at the top queue.push(make_pair((p.first + cost[make_pair(p.second, graph[p.second][i])]) * -1, graph[p.second][i])); } // mark as visited visited[p.second] = 1; } return answer; } // main function int main() { // create the graph graph.resize(7); // add edge graph[0].push_back(1); graph[0].push_back(3); graph[3].push_back(1); graph[3].push_back(6); graph[3].push_back(4); graph[1].push_back(6); graph[4].push_back(2); graph[4].push_back(5); graph[2].push_back(1); graph[5].push_back(2); graph[5].push_back(6); graph[6].push_back(4); // add the cost cost[make_pair(0, 1)] = 2; cost[make_pair(0, 3)] = 5; cost[make_pair(1, 6)] = 1; cost[make_pair(3, 1)] = 5; cost[make_pair(3, 6)] = 6; cost[make_pair(3, 4)] = 2; cost[make_pair(2, 1)] = 4; cost[make_pair(4, 2)] = 4; cost[make_pair(4, 5)] = 3; cost[make_pair(5, 2)] = 6; cost[make_pair(5, 6)] = 3; cost[make_pair(6, 4)] = 7; // goal state vector<int> goal; // set the goal // there can be multiple goal states goal.push_back(6); // get the answer vector<int> answer = uniform_cost_search(goal, 0); // print the answer cout << "Minimum cost from 0 to 6 is = " << answer[0] << endl; return 0; }
Python3
# Python3 implementation of above approach # returns the minimum cost in a vector( if # there are multiple goal states) def uniform_cost_search(goal, start): # minimum cost upto # goal state from starting global graph,cost answer = [] # create a priority queue queue = [] # set the answer vector to max value for i in range(len(goal)): answer.append(10**8) # insert the starting index queue.append([0, start]) # map to store visited node visited = {} # count count = 0 # while the queue is not empty while (len(queue) > 0): # get the top element of the queue = sorted(queue) p = queue[-1] # pop the element del queue[-1] # get the original value p[0] *= -1 # check if the element is part of # the goal list if (p[1] in goal): # get the position index = goal.index(p[1]) # if a new goal is reached if (answer[index] == 10**8): count += 1 # if the cost is less if (answer[index] > p[0]): answer[index] = p[0] # pop the element del queue[-1] queue = sorted(queue) if (count == len(goal)): return answer # check for the non visited nodes # which are adjacent to present node if (p[1] not in visited): for i in range(len(graph[p[1]])): # value is multiplied by -1 so that # least priority is at the top queue.append( [(p[0] + cost[(p[1], graph[p[1]][i])])* -1, graph[p[1]][i]]) # mark as visited visited[p[1]] = 1 return answer # main function if __name__ == '__main__': # create the graph graph,cost = [[] for i in range(8)],{} # add edge graph[0].append(1) graph[0].append(3) graph[3].append(1) graph[3].append(6) graph[3].append(4) graph[1].append(6) graph[4].append(2) graph[4].append(5) graph[2].append(1) graph[5].append(2) graph[5].append(6) graph[6].append(4) # add the cost cost[(0, 1)] = 2 cost[(0, 3)] = 5 cost[(1, 6)] = 1 cost[(3, 1)] = 5 cost[(3, 6)] = 6 cost[(3, 4)] = 2 cost[(2, 1)] = 4 cost[(4, 2)] = 4 cost[(4, 5)] = 3 cost[(5, 2)] = 6 cost[(5, 6)] = 3 cost[(6, 4)] = 7 # goal state goal = [] # set the goal # there can be multiple goal states goal.append(6) # get the answer answer = uniform_cost_search(goal, 0) # print the answer print("Minimum cost from 0 to 6 is = ",answer[0]) # This code is contributed by mohit kumar 29
C#
// C# implementation of above approach using System; using System.Collections; using System.Collections.Generic; class GFG { // graph static List<List<int>> graph=new List<List<int>>(); // map to store cost of edges static Dictionary<Tuple<int,int>,int> cost= new Dictionary<Tuple<int,int>,int>(); // returns the minimum cost in a vector( if // there are multiple goal states) static List<int> uniform_cost_search(List<int> goal, int start) { // minimum cost upto // goal state from starting // state List<int> answer=new List<int>(); // create a priority queue List<Tuple<int, int> > queue = new List<Tuple<int, int> >(); // set the answer vector to max value for (int i = 0; i < goal.Count; i++) answer.Add(int.MaxValue); // insert the starting index queue.Add(new Tuple<int,int>(0, start)); // map to store visited node Dictionary<int, int> visited=new Dictionary<int,int>(); // count int count = 0; // while the queue is not empty while (queue.Count > 0) { // get the top element of the // priority queue Tuple<int, int> q = queue[0]; Tuple<int, int> p = new Tuple<int,int>(-q.Item1,q.Item2); // pop the element queue.RemoveAt(0); // check if the element is part of // the goal list if (goal.Contains(p.Item2)) { // get the position int index = goal.IndexOf(p.Item2); // if a new goal is reached if (answer[index] == int.MaxValue) count++; // if the cost is less if (answer[index] > p.Item1) answer[index] = p.Item1; // pop the element queue.RemoveAt(0); // if all goals are reached if (count == goal.Count) return answer; } // check for the non visited nodes // which are adjacent to present node if (!visited.ContainsKey(p.Item2)) for (int i = 0; i < graph[p.Item2].Count; i++) { // value is multiplied by -1 so that // least priority is at the top queue.Add(new Tuple<int,int>((p.Item1 + (cost.ContainsKey(new Tuple<int,int>(p.Item2, graph[p.Item2][i])) ? cost[new Tuple<int,int>(p.Item2, graph[p.Item2][i])] : 0))*-1, graph[p.Item2][i])); } // mark as visited visited[p.Item2] = 1; } return answer; } // main function public static void Main(params string []args) { // create the graph graph=new List<List<int>>(); for(int i=0;i<7;i++) { graph.Add(new List<int>()); } // add edge graph[0].Add(1); graph[0].Add(3); graph[3].Add(1); graph[3].Add(6); graph[3].Add(4); graph[1].Add(6); graph[4].Add(2); graph[4].Add(5); graph[2].Add(1); graph[5].Add(2); graph[5].Add(6); graph[6].Add(4); // add the cost cost[new Tuple<int,int>(0, 1)] = 2; cost[new Tuple<int,int>(0, 3)] = 5; cost[new Tuple<int,int>(1, 6)] = 1; cost[new Tuple<int,int>(3, 1)] = 5; cost[new Tuple<int,int>(3, 6)] = 6; cost[new Tuple<int,int>(3, 4)] = 2; cost[new Tuple<int,int>(2, 1)] = 4; cost[new Tuple<int,int>(4, 2)] = 4; cost[new Tuple<int,int>(4, 5)] = 3; cost[new Tuple<int,int>(5, 2)] = 6; cost[new Tuple<int,int>(5, 6)] = 3; cost[new Tuple<int,int>(6, 4)] = 7; // goal state List<int> goal=new List<int>(); // set the goal // there can be multiple goal states goal.Add(6); // get the answer List<int> answer = uniform_cost_search(goal, 0); // print the answer Console.Write("Minimum cost from 0 to 6 is = " + answer[0]); } } // This code is contributed by rutvik_56.
Minimum cost from 0 to 6 is = 3
Complejidad: O( m ^ (1+piso(l/e)))
donde,
m es el número máximo de vecinos que tiene un Node
l es la longitud del camino más corto hacia el estado objetivo
e es el menor costo de un borde
Publicación traducida automáticamente
Artículo escrito por andrew1234 y traducido por Barcelona Geeks. The original can be accessed here. Licence: CCBY-SA