ハミルトニアンサイクルとは何ですか?
ハミルトニアン サイクルまたは回路 グラフで G のすべての頂点を訪れるサイクルです。 G 正確に 1 回だけ実行し、開始頂点に戻ります。
- グラフにハミルトニアン サイクルが含まれる場合、それは次のように呼ばれます。 ハミルトニアングラフ それ以外の場合はそうです 非ハミルトニアン 。
- グラフ内でハミルトニアン サイクルを見つけることはよく知られています。 NP完全問題 これは、すべての種類のグラフに対してそれを解決する既知の効率的なアルゴリズムが存在しないことを意味します。ただし、小さいグラフや特定の種類のグラフの場合は解決できます。
ハミルトニアン サイクル問題は、次のようなさまざまな分野で実用化されています。 物流、ネットワーク設計、コンピューターサイエンス 。
ハミルトニアンパスとは何ですか?
ハミルトニアン パス グラフで G は G のすべての頂点を 1 回だけ訪問するパスであり、 ハミルトニアン パス 開始頂点に戻る必要はありません。それは開かれた道です。
- に似ています ハミルトニアンサイクル 問題、見つける ハミルトニアン パス 一般的なグラフでも NP完全 挑戦的になる可能性があります。ただし、多くの場合、ハミルトニアン サイクルを見つけるよりも簡単な問題です。
- ハミルトニアン パスは、次のようなさまざまな分野で応用できます。 交通ネットワークにおける最適ルートの探索、回路設計、グラフ理論の研究 。
問題の声明: 無向グラフが与えられた場合、タスクは、グラフにハミルトニアン サイクルが含まれているかどうかを判断することです。含まれている場合は、パスを出力します。
例:
推奨: で解決してください 練習する 解決策に進む前に、まず。入力: グラフ[][] = {{0, 1, 0, 1, 0},{1, 0, 1, 1, 1},{0, 1, 0, 0, 1},{1, 1, 0, 0, 1},{0, 1, 1, 1, 0}}
入力グラフ[][]
文字列Javaで分割出力: {0、1、2、4、3、0}。
入力: グラフ[][] = {{0, 1, 0, 1, 0},{1, 0, 1, 1, 1},{0, 1, 0, 0, 1},{1, 1, 0, 0, 0},{0, 1, 1, 0, 0}}
入力グラフ[][]
出力: 解決策は存在しません
単純なアルゴリズム : この問題は、以下のアイデアを使用して解決できます。
考えられる頂点の構成をすべて生成し、指定された制約を満たす構成を出力します。ンもあるでしょう! (n 階乗) 構成。したがって、このアプローチの全体的な時間計算量は次のようになります。 の上!)。
ハミルトニアンサイクルを使用する バックトラッキングアルゴリズム :
空のパス配列を作成し、頂点を追加します 0 それに。頂点を起点として他の頂点を追加します 1 。頂点を追加する前に、その頂点が以前に追加した頂点に隣接していて、まだ追加されていないことを確認してください。そのような頂点が見つかった場合は、その頂点をソリューションの一部として追加します。頂点が見つからない場合は戻ります 間違い 。
イラスト:
次のグラフのハミルトニアン サイクルを調べてみましょう。
- ノード 0 から始めます。
- DFS を適用してハミルトニアン パスを見つけます。
- 基本ケースに到達したとき(すなわち、 通過したノードの総数 == V (頂点の総数) ):
- 現在のノードが開始ノードの隣接ノードであるかどうかを確認します。
- ノードとして 2 とノード 0 お互いに隣人ではないので、そこから戻ります。
開始ノード 0 から開始して DFS を呼び出す
- パス {0, 3, 1, 4, 2} にサイクルが見つからないため。ということで、ノード2、ノード4から戻ります。
- ここで、ノード 1 (つまりノード 2) の別のオプションを検討します。
- 再びベース条件に達したら、ハミルトニアン サイクルを確認します。
- ノード 4 はノード 0 の隣接ノードではないため、再びサイクルが見つからず、リターンします。
- ノード 4、ノード 2、ノード 1 から戻ります。
- 次に、ノード 3 の他のオプションを検討します。
ハミルトニアンサイクル
- ハミルトニアンパスで {0,3,4,2,1,0} ノード 1 がノード 0 の隣接ノードであるため、サイクルが得られます。
- したがって、この循環パスを出力します。
- これがハミルトニアン サイクルです。
以下は、ハミルトニアン サイクルを見つけるためのバックトラッキングの実装です。
C++ /* C++ program for solution of Hamiltonian Cycle problem using backtracking */ #include using namespace std; // Number of vertices in the graph #define V 5 void printSolution(int path[]); /* A utility function to check if the vertex v can be added at index 'pos' in the Hamiltonian Cycle constructed so far (stored in 'path[]') */ bool isSafe(int v, bool graph[V][V], int path[], int pos) { /* Check if this vertex is an adjacent vertex of the previously added vertex. */ if (graph [path[pos - 1]][ v ] == 0) return false; /* Check if the vertex has already been included. This step can be optimized by creating an array of size V */ for (int i = 0; i < pos; i++) if (path[i] == v) return false; return true; } /* A recursive utility function to solve hamiltonian cycle problem */ bool hamCycleUtil(bool graph[V][V], int path[], int pos) { /* base case: If all vertices are included in Hamiltonian Cycle */ if (pos == V) { // And if there is an edge from the // last included vertex to the first vertex if (graph[path[pos - 1]][path[0]] == 1) return true; else return false; } // Try different vertices as a next candidate // in Hamiltonian Cycle. We don't try for 0 as // we included 0 as starting point in hamCycle() for (int v = 1; v < V; v++) { /* Check if this vertex can be added // to Hamiltonian Cycle */ if (isSafe(v, graph, path, pos)) { path[pos] = v; /* recur to construct rest of the path */ if (hamCycleUtil (graph, path, pos + 1) == true) return true; /* If adding vertex v doesn't lead to a solution, then remove it */ path[pos] = -1; } } /* If no vertex can be added to Hamiltonian Cycle constructed so far, then return false */ return false; } /* This function solves the Hamiltonian Cycle problem using Backtracking. It mainly uses hamCycleUtil() to solve the problem. It returns false if there is no Hamiltonian Cycle possible, otherwise return true and prints the path. Please note that there may be more than one solutions, this function prints one of the feasible solutions. */ bool hamCycle(bool graph[V][V]) { int *path = new int[V]; for (int i = 0; i < V; i++) path[i] = -1; /* Let us put vertex 0 as the first vertex in the path. If there is a Hamiltonian Cycle, then the path can be started from any point of the cycle as the graph is undirected */ path[0] = 0; if (hamCycleUtil(graph, path, 1) == false ) { cout << '
Solution does not exist'; return false; } printSolution(path); return true; } /* A utility function to print solution */ void printSolution(int path[]) { cout << 'Solution Exists:' ' Following is one Hamiltonian Cycle
'; for (int i = 0; i < V; i++) cout << path[i] << ' '; // Let us print the first vertex again // to show the complete cycle cout << path[0] << ' '; cout << endl; } // Driver Code int main() { /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3)-------(4) */ bool graph1[V][V] = {{0, 1, 0, 1, 0}, {1, 0, 1, 1, 1}, {0, 1, 0, 0, 1}, {1, 1, 0, 0, 1}, {0, 1, 1, 1, 0}}; // Print the solution hamCycle(graph1); /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3) (4) */ bool graph2[V][V] = {{0, 1, 0, 1, 0}, {1, 0, 1, 1, 1}, {0, 1, 0, 0, 1}, {1, 1, 0, 0, 0}, {0, 1, 1, 0, 0}}; // Print the solution hamCycle(graph2); return 0; } // This is code is contributed by rathbhupendra> C++ #include using namespace std; int main() { cout << 'GFG!'; return 0; }> C /* C program for solution of Hamiltonian Cycle problem using backtracking */ #include // Number of vertices in the graph #define V 5 void printSolution(int path[]); /* A utility function to check if the vertex v can be added at index 'pos' in the Hamiltonian Cycle constructed so far (stored in 'path[]') */ int isSafe(int v, int graph[V][V], int path[], int pos) { /* Check if this vertex is an adjacent vertex of the previously added vertex. */ if (graph [ path[pos-1] ][ v ] == 0) return 0; /* Check if the vertex has already been included. This step can be optimized by creating an array of size V */ for (int i = 0; i < pos; i++) if (path[i] == v) return 0; return 1; } /* A recursive utility function to solve hamiltonian cycle problem */ int hamCycleUtil(int graph[V][V], int path[], int pos) { /* base case: If all vertices are included in Hamiltonian Cycle */ if (pos == V) { // And if there is an edge from the last included vertex to the // first vertex if ( graph[ path[pos-1] ][ path[0] ] == 1 ) return 1; else return 0; } // Try different vertices as a next candidate in Hamiltonian Cycle. // We don't try for 0 as we included 0 as starting point in hamCycle() for (int v = 1; v < V; v++) { /* Check if this vertex can be added to Hamiltonian Cycle */ if (isSafe(v, graph, path, pos)) { path[pos] = v; /* recur to construct rest of the path */ if (hamCycleUtil (graph, path, pos+1) == 1) return 1; /* If adding vertex v doesn't lead to a solution, then remove it */ path[pos] = -1; } } /* If no vertex can be added to Hamiltonian Cycle constructed so far, then return false */ return 0; } /* This function solves the Hamiltonian Cycle problem using Backtracking. It mainly uses hamCycleUtil() to solve the problem. It returns false if there is no Hamiltonian Cycle possible, otherwise return true and prints the path. Please note that there may be more than one solutions, this function prints one of the feasible solutions. */ int hamCycle(int graph[V][V]) { int path[V]; for (int i = 0; i < V; i++) path[i] = -1; /* Let us put vertex 0 as the first vertex in the path. If there is a Hamiltonian Cycle, then the path can be started from any point of the cycle as the graph is undirected */ path[0] = 0; if ( hamCycleUtil(graph, path, 1) == 0 ) { printf('
Solution does not exist'); return 0; } printSolution(path); return 1; } /* A utility function to print solution */ void printSolution(int path[]) { printf ('Solution Exists:' ' Following is one Hamiltonian Cycle
'); for (int i = 0; i < V; i++) printf(' %d ', path[i]); // Let us print the first vertex again to show the complete cycle printf(' %d ', path[0]); printf('
'); } // driver program to test above function int main() { /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3)-------(4) */ int graph1[V][V] = {{0, 1, 0, 1, 0}, {1, 0, 1, 1, 1}, {0, 1, 0, 0, 1}, {1, 1, 0, 0, 1}, {0, 1, 1, 1, 0}, }; // Print the solution hamCycle(graph1); /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3) (4) */ int graph2[V][V] = {{0, 1, 0, 1, 0}, {1, 0, 1, 1, 1}, {0, 1, 0, 0, 1}, {1, 1, 0, 0, 0}, {0, 1, 1, 0, 0}, }; // Print the solution hamCycle(graph2); return 0; }> ジャワ /* Java program for solution of Hamiltonian Cycle problem using backtracking */ class HamiltonianCycle { final int V = 5; int path[]; /* A utility function to check if the vertex v can be added at index 'pos'in the Hamiltonian Cycle constructed so far (stored in 'path[]') */ boolean isSafe(int v, int graph[][], int path[], int pos) { /* Check if this vertex is an adjacent vertex of the previously added vertex. */ if (graph[path[pos - 1]][v] == 0) return false; /* Check if the vertex has already been included. This step can be optimized by creating an array of size V */ for (int i = 0; i < pos; i++) if (path[i] == v) return false; return true; } /* A recursive utility function to solve hamiltonian cycle problem */ boolean hamCycleUtil(int graph[][], int path[], int pos) { /* base case: If all vertices are included in Hamiltonian Cycle */ if (pos == V) { // And if there is an edge from the last included // vertex to the first vertex if (graph[path[pos - 1]][path[0]] == 1) return true; else return false; } // Try different vertices as a next candidate in // Hamiltonian Cycle. We don't try for 0 as we // included 0 as starting point in hamCycle() for (int v = 1; v < V; v++) { /* Check if this vertex can be added to Hamiltonian Cycle */ if (isSafe(v, graph, path, pos)) { path[pos] = v; /* recur to construct rest of the path */ if (hamCycleUtil(graph, path, pos + 1) == true) return true; /* If adding vertex v doesn't lead to a solution, then remove it */ path[pos] = -1; } } /* If no vertex can be added to Hamiltonian Cycle constructed so far, then return false */ return false; } /* This function solves the Hamiltonian Cycle problem using Backtracking. It mainly uses hamCycleUtil() to solve the problem. It returns false if there is no Hamiltonian Cycle possible, otherwise return true and prints the path. Please note that there may be more than one solutions, this function prints one of the feasible solutions. */ int hamCycle(int graph[][]) { path = new int[V]; for (int i = 0; i < V; i++) path[i] = -1; /* Let us put vertex 0 as the first vertex in the path. If there is a Hamiltonian Cycle, then the path can be started from any point of the cycle as the graph is undirected */ path[0] = 0; if (hamCycleUtil(graph, path, 1) == false) { System.out.println('
Solution does not exist'); return 0; } printSolution(path); return 1; } /* A utility function to print solution */ void printSolution(int path[]) { System.out.println('Solution Exists: Following' + ' is one Hamiltonian Cycle'); for (int i = 0; i < V; i++) System.out.print(' ' + path[i] + ' '); // Let us print the first vertex again to show the // complete cycle System.out.println(' ' + path[0] + ' '); } // driver program to test above function public static void main(String args[]) { HamiltonianCycle hamiltonian = new HamiltonianCycle(); /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3)-------(4) */ int graph1[][] = {{0, 1, 0, 1, 0}, {1, 0, 1, 1, 1}, {0, 1, 0, 0, 1}, {1, 1, 0, 0, 1}, {0, 1, 1, 1, 0}, }; // Print the solution hamiltonian.hamCycle(graph1); /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3) (4) */ int graph2[][] = {{0, 1, 0, 1, 0}, {1, 0, 1, 1, 1}, {0, 1, 0, 0, 1}, {1, 1, 0, 0, 0}, {0, 1, 1, 0, 0}, }; // Print the solution hamiltonian.hamCycle(graph2); } } // This code is contributed by Abhishek Shankhadhar> パイソン # Python program for solution of # hamiltonian cycle problem class Graph(): def __init__(self, vertices): self.graph = [[0 for column in range(vertices)] for row in range(vertices)] self.V = vertices ''' Check if this vertex is an adjacent vertex of the previously added vertex and is not included in the path earlier ''' def isSafe(self, v, pos, path): # Check if current vertex and last vertex # in path are adjacent if self.graph[ path[pos-1] ][v] == 0: return False # Check if current vertex not already in path for vertex in path: if vertex == v: return False return True # A recursive utility function to solve # hamiltonian cycle problem def hamCycleUtil(self, path, pos): # base case: if all vertices are # included in the path if pos == self.V: # Last vertex must be adjacent to the # first vertex in path to make a cycle if self.graph[ path[pos-1] ][ path[0] ] == 1: return True else: return False # Try different vertices as a next candidate # in Hamiltonian Cycle. We don't try for 0 as # we included 0 as starting point in hamCycle() for v in range(1,self.V): if self.isSafe(v, pos, path) == True: path[pos] = v if self.hamCycleUtil(path, pos+1) == True: return True # Remove current vertex if it doesn't # lead to a solution path[pos] = -1 return False def hamCycle(self): path = [-1] * self.V ''' Let us put vertex 0 as the first vertex in the path. If there is a Hamiltonian Cycle, then the path can be started from any point of the cycle as the graph is undirected ''' path[0] = 0 if self.hamCycleUtil(path,1) == False: print ('Solution does not exist
') return False self.printSolution(path) return True def printSolution(self, path): print ('Solution Exists: Following', 'is one Hamiltonian Cycle') for vertex in path: print (vertex ) # Driver Code ''' Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3)-------(4) ''' g1 = Graph(5) g1.graph = [ [0, 1, 0, 1, 0], [1, 0, 1, 1, 1], [0, 1, 0, 0, 1,],[1, 1, 0, 0, 1], [0, 1, 1, 1, 0], ] # Print the solution g1.hamCycle(); ''' Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3) (4) ''' g2 = Graph(5) g2.graph = [ [0, 1, 0, 1, 0], [1, 0, 1, 1, 1], [0, 1, 0, 0, 1,], [1, 1, 0, 0, 0], [0, 1, 1, 0, 0], ] # Print the solution g2.hamCycle(); # This code is contributed by Divyanshu Mehta> C# // C# program for solution of Hamiltonian // Cycle problem using backtracking using System; public class HamiltonianCycle { readonly int V = 5; int []path; /* A utility function to check if the vertex v can be added at index 'pos'in the Hamiltonian Cycle constructed so far (stored in 'path[]') */ bool isSafe(int v, int [,]graph, int []path, int pos) { /* Check if this vertex is an adjacent vertex of the previously added vertex. */ if (graph[path[pos - 1], v] == 0) return false; /* Check if the vertex has already been included. This step can be optimized by creating an array of size V */ for (int i = 0; i < pos; i++) if (path[i] == v) return false; return true; } /* A recursive utility function to solve hamiltonian cycle problem */ bool hamCycleUtil(int [,]graph, int []path, int pos) { /* base case: If all vertices are included in Hamiltonian Cycle */ if (pos == V) { // And if there is an edge from the last included // vertex to the first vertex if (graph[path[pos - 1],path[0]] == 1) return true; else return false; } // Try different vertices as a next candidate in // Hamiltonian Cycle. We don't try for 0 as we // included 0 as starting point in hamCycle() for (int v = 1; v < V; v++) { /* Check if this vertex can be added to Hamiltonian Cycle */ if (isSafe(v, graph, path, pos)) { path[pos] = v; /* recur to construct rest of the path */ if (hamCycleUtil(graph, path, pos + 1) == true) return true; /* If adding vertex v doesn't lead to a solution, then remove it */ path[pos] = -1; } } /* If no vertex can be added to Hamiltonian Cycle constructed so far, then return false */ return false; } /* This function solves the Hamiltonian Cycle problem using Backtracking. It mainly uses hamCycleUtil() to solve the problem. It returns false if there is no Hamiltonian Cycle possible, otherwise return true and prints the path. Please note that there may be more than one solutions, this function prints one of the feasible solutions. */ int hamCycle(int [,]graph) { path = new int[V]; for (int i = 0; i < V; i++) path[i] = -1; /* Let us put vertex 0 as the first vertex in the path. If there is a Hamiltonian Cycle, then the path can be started from any point of the cycle as the graph is undirected */ path[0] = 0; if (hamCycleUtil(graph, path, 1) == false) { Console.WriteLine('
Solution does not exist'); return 0; } printSolution(path); return 1; } /* A utility function to print solution */ void printSolution(int []path) { Console.WriteLine('Solution Exists: Following' + ' is one Hamiltonian Cycle'); for (int i = 0; i < V; i++) Console.Write(' ' + path[i] + ' '); // Let us print the first vertex again // to show the complete cycle Console.WriteLine(' ' + path[0] + ' '); } // Driver code public static void Main(String []args) { HamiltonianCycle hamiltonian = new HamiltonianCycle(); /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3)-------(4) */ int [,]graph1= {{0, 1, 0, 1, 0}, {1, 0, 1, 1, 1}, {0, 1, 0, 0, 1}, {1, 1, 0, 0, 1}, {0, 1, 1, 1, 0}, }; // Print the solution hamiltonian.hamCycle(graph1); /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3) (4) */ int [,]graph2 = {{0, 1, 0, 1, 0}, {1, 0, 1, 1, 1}, {0, 1, 0, 0, 1}, {1, 1, 0, 0, 0}, {0, 1, 1, 0, 0}, }; // Print the solution hamiltonian.hamCycle(graph2); } } // This code contributed by Rajput-Ji> JavaScript >>
PHP // PHP program for solution of // Hamiltonian Cycle problem // using backtracking $V = 5; /* A utility function to check if the vertex v can be added at index 'pos' in the Hamiltonian Cycle constructed so far (stored in 'path[]') */ function isSafe($v, $graph, &$path, $pos) { /* Check if this vertex is an adjacent vertex of the previously added vertex. */ if ($graph[$path[$pos - 1]][$v] == 0) return false; /* Check if the vertex has already been included. This step can be optimized by creating an array of size V */ for ($i = 0; $i < $pos; $i++) if ($path[$i] == $v) return false; return true; } /* A recursive utility function to solve hamiltonian cycle problem */ function hamCycleUtil($graph, &$path, $pos) { global $V; /* base case: If all vertices are included in Hamiltonian Cycle */ if ($pos == $V) { // And if there is an edge from the // last included vertex to the first vertex if ($graph[$path[$pos - 1]][$path[0]] == 1) return true; else return false; } // Try different vertices as a next candidate in // Hamiltonian Cycle. We don't try for 0 as we // included 0 as starting point hamCycle() for ($v = 1; $v < $V; $v++) { /* Check if this vertex can be added to Hamiltonian Cycle */ if (isSafe($v, $graph, $path, $pos)) { $path[$pos] = $v; /* recur to construct rest of the path */ if (hamCycleUtil($graph, $path, $pos + 1) == true) return true; /* If adding vertex v doesn't lead to a solution, then remove it */ $path[$pos] = -1; } } /* If no vertex can be added to Hamiltonian Cycle constructed so far, then return false */ return false; } /* This function solves the Hamiltonian Cycle problem using Backtracking. It mainly uses hamCycleUtil() to solve the problem. It returns false if there is no Hamiltonian Cycle possible, otherwise return true and prints the path. Please note that there may be more than one solutions, this function prints one of the feasible solutions. */ function hamCycle($graph) { global $V; $path = array_fill(0, $V, 0); for ($i = 0; $i < $V; $i++) $path[$i] = -1; /* Let us put vertex 0 as the first vertex in the path. If there is a Hamiltonian Cycle, then the path can be started from any point of the cycle as the graph is undirected */ $path[0] = 0; if (hamCycleUtil($graph, $path, 1) == false) { echo('
Solution does not exist'); return 0; } printSolution($path); return 1; } /* A utility function to print solution */ function printSolution($path) { global $V; echo('Solution Exists: Following is '. 'one Hamiltonian Cycle
'); for ($i = 0; $i < $V; $i++) echo(' '.$path[$i].' '); // Let us print the first vertex again to show the // complete cycle echo(' '.$path[0].'
'); } // Driver Code /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3)-------(4) */ $graph1 = array(array(0, 1, 0, 1, 0), array(1, 0, 1, 1, 1), array(0, 1, 0, 0, 1), array(1, 1, 0, 0, 1), array(0, 1, 1, 1, 0), ); // Print the solution hamCycle($graph1); /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3) (4) */ $graph2 = array(array(0, 1, 0, 1, 0), array(1, 0, 1, 1, 1), array(0, 1, 0, 0, 1), array(1, 1, 0, 0, 0), array(0, 1, 1, 0, 0)); // Print the solution hamCycle($graph2); // This code is contributed by mits ?>>> 出力
Solution Exists: Following is one Hamiltonian Cycle 0 1 2 4 3 0 Solution does not exist>
時間計算量 : O(N!)、 ここで、N は頂点の数です。
補助スペース : O(1)、 余分なスペースが使用されないため。
注記: 上記のコードは常に次から始まるサイクルを出力します。 0 。サイクルはどの時点からでも開始できるため、開始点は重要ではありません。開始点を変更したい場合は、上記のコードに 2 つの変更を加える必要があります。
パス[0] = 0を変更します。に パス[0] = s ;どこ s あなたの新しいです 出発点 。また、ループを (int v = 1; v に変更します)




