我有一个nxm个由非负整数组成的矩阵。例如:
2 3 4 7 1
1 5 2 6 2
4 3 4 2 1
2 1 2 4 1
3 1 3 4 1
2 1 4 3 2
6 9 1 6 4
“投下炸弹”会使目标细胞及其所有八个邻居的数量减少一个，直到最小值为零。
x x x
x X x
x x x
什么样的算法可以确定将所有细胞减少到零所需的最少炸弹数量?
B选项(因为我不是一个细心的读者)
事实上，问题的第一个版本并不是我要寻找的答案。我没有仔细阅读整个任务，有额外的约束条件，让我们说:
那么简单的问题是，当行中的序列必须是非递增的:
8 7 6 6 5是可能的输入序列
7 8 5 5 2是不可能的，因为7 -> 8在一个序列中增长。
也许为“简单”的问题找到答案会有助于为更难的问题找到解决方案。
PS:我相信当我们有几个相同的情况需要最少的炸弹来清除上面的线时，我们会选择在“左侧”使用最多炸弹的一个。还有什么证据是正确的吗?
当前回答
我想不出一个计算实际数字的方法除非用我最好的启发式方法计算轰炸行动并希望得到一个合理的结果。
So my method is to compute a bombing efficiency metric for each cell, bomb the cell with the highest value, .... iterate the process until I've flattened everything. Some have advocated using simple potential damage (i.e. score from 0 to 9) as a metric, but that falls short by pounding high value cells and not making use of damage overlap. I'd calculate cell value - sum of all neighbouring cells, reset any positive to 0 and use the absolute value of anything negative. Intuitively this metric should make a selection that help maximise damage overlap on cells with high counts instead of pounding those directly.
下面的代码在28个炸弹中达到了测试场的完全破坏(注意，使用潜在伤害作为度量，结果是31!)
using System;
using System.Collections.Generic;
using System.Linq;
namespace StackOverflow
{
internal class Program
{
// store the battle field as flat array + dimensions
private static int _width = 5;
private static int _length = 7;
private static int[] _field = new int[] {
2, 3, 4, 7, 1,
1, 5, 2, 6, 2,
4, 3, 4, 2, 1,
2, 1, 2, 4, 1,
3, 1, 3, 4, 1,
2, 1, 4, 3, 2,
6, 9, 1, 6, 4
};
// this will store the devastation metric
private static int[] _metric;
// do the work
private static void Main(string[] args)
{
int count = 0;
while (_field.Sum() > 0)
{
Console.Out.WriteLine("Round {0}:", ++count);
GetBlastPotential();
int cell_to_bomb = FindBestBombingSite();
PrintField(cell_to_bomb);
Bomb(cell_to_bomb);
}
Console.Out.WriteLine("Done in {0} rounds", count);
}
// convert 2D position to 1D index
private static int Get1DCoord(int x, int y)
{
if ((x < 0) || (y < 0) || (x >= _width) || (y >= _length)) return -1;
else
{
return (y * _width) + x;
}
}
// Convert 1D index to 2D position
private static void Get2DCoord(int n, out int x, out int y)
{
if ((n < 0) || (n >= _field.Length))
{
x = -1;
y = -1;
}
else
{
x = n % _width;
y = n / _width;
}
}
// Compute a list of 1D indices for a cell neighbours
private static List<int> GetNeighbours(int cell)
{
List<int> neighbours = new List<int>();
int x, y;
Get2DCoord(cell, out x, out y);
if ((x >= 0) && (y >= 0))
{
List<int> tmp = new List<int>();
tmp.Add(Get1DCoord(x - 1, y - 1));
tmp.Add(Get1DCoord(x - 1, y));
tmp.Add(Get1DCoord(x - 1, y + 1));
tmp.Add(Get1DCoord(x, y - 1));
tmp.Add(Get1DCoord(x, y + 1));
tmp.Add(Get1DCoord(x + 1, y - 1));
tmp.Add(Get1DCoord(x + 1, y));
tmp.Add(Get1DCoord(x + 1, y + 1));
// eliminate invalid coords - i.e. stuff past the edges
foreach (int c in tmp) if (c >= 0) neighbours.Add(c);
}
return neighbours;
}
// Compute the devastation metric for each cell
// Represent the Value of the cell minus the sum of all its neighbours
private static void GetBlastPotential()
{
_metric = new int[_field.Length];
for (int i = 0; i < _field.Length; i++)
{
_metric[i] = _field[i];
List<int> neighbours = GetNeighbours(i);
if (neighbours != null)
{
foreach (int j in neighbours) _metric[i] -= _field[j];
}
}
for (int i = 0; i < _metric.Length; i++)
{
_metric[i] = (_metric[i] < 0) ? Math.Abs(_metric[i]) : 0;
}
}
//// Compute the simple expected damage a bomb would score
//private static void GetBlastPotential()
//{
// _metric = new int[_field.Length];
// for (int i = 0; i < _field.Length; i++)
// {
// _metric[i] = (_field[i] > 0) ? 1 : 0;
// List<int> neighbours = GetNeighbours(i);
// if (neighbours != null)
// {
// foreach (int j in neighbours) _metric[i] += (_field[j] > 0) ? 1 : 0;
// }
// }
//}
// Update the battle field upon dropping a bomb
private static void Bomb(int cell)
{
List<int> neighbours = GetNeighbours(cell);
foreach (int i in neighbours)
{
if (_field[i] > 0) _field[i]--;
}
}
// Find the best bombing site - just return index of local maxima
private static int FindBestBombingSite()
{
int max_idx = 0;
int max_val = int.MinValue;
for (int i = 0; i < _metric.Length; i++)
{
if (_metric[i] > max_val)
{
max_val = _metric[i];
max_idx = i;
}
}
return max_idx;
}
// Display the battle field on the console
private static void PrintField(int cell)
{
for (int x = 0; x < _width; x++)
{
for (int y = 0; y < _length; y++)
{
int c = Get1DCoord(x, y);
if (c == cell)
Console.Out.Write(string.Format("[{0}]", _field[c]).PadLeft(4));
else
Console.Out.Write(string.Format(" {0} ", _field[c]).PadLeft(4));
}
Console.Out.Write(" || ");
for (int y = 0; y < _length; y++)
{
int c = Get1DCoord(x, y);
if (c == cell)
Console.Out.Write(string.Format("[{0}]", _metric[c]).PadLeft(4));
else
Console.Out.Write(string.Format(" {0} ", _metric[c]).PadLeft(4));
}
Console.Out.WriteLine();
}
Console.Out.WriteLine();
}
}
}
产生的轰炸模式输出如下(左边是字段值，右边是度量值)
Round 1:
2 1 4 2 3 2 6 || 7 16 8 10 4 18 6
3 5 3 1 1 1 9 || 11 18 18 21 17 28 5
4 [2] 4 2 3 4 1 || 19 [32] 21 20 17 24 22
7 6 2 4 4 3 6 || 8 17 20 14 16 22 8
1 2 1 1 1 2 4 || 14 15 14 11 13 16 7
Round 2:
2 1 4 2 3 2 6 || 5 13 6 9 4 18 6
2 4 2 1 1 [1] 9 || 10 15 17 19 17 [28] 5
3 2 3 2 3 4 1 || 16 24 18 17 17 24 22
6 5 1 4 4 3 6 || 7 14 19 12 16 22 8
1 2 1 1 1 2 4 || 12 12 12 10 13 16 7
Round 3:
2 1 4 2 2 1 5 || 5 13 6 7 3 15 5
2 4 2 1 0 1 8 || 10 15 17 16 14 20 2
3 [2] 3 2 2 3 0 || 16 [24] 18 15 16 21 21
6 5 1 4 4 3 6 || 7 14 19 11 14 19 6
1 2 1 1 1 2 4 || 12 12 12 10 13 16 7
Round 4:
2 1 4 2 2 1 5 || 3 10 4 6 3 15 5
1 3 1 1 0 1 8 || 9 12 16 14 14 20 2
2 2 2 2 2 [3] 0 || 13 16 15 12 16 [21] 21
5 4 0 4 4 3 6 || 6 11 18 9 14 19 6
1 2 1 1 1 2 4 || 10 9 10 9 13 16 7
Round 5:
2 1 4 2 2 1 5 || 3 10 4 6 2 13 3
1 3 1 1 0 [0] 7 || 9 12 16 13 12 [19] 2
2 2 2 2 1 3 0 || 13 16 15 10 14 15 17
5 4 0 4 3 2 5 || 6 11 18 7 13 17 6
1 2 1 1 1 2 4 || 10 9 10 8 11 13 5
Round 6:
2 1 4 2 1 0 4 || 3 10 4 5 2 11 2
1 3 1 1 0 0 6 || 9 12 16 11 8 13 0
2 2 2 2 0 2 0 || 13 16 15 9 14 14 15
5 4 [0] 4 3 2 5 || 6 11 [18] 6 11 15 5
1 2 1 1 1 2 4 || 10 9 10 8 11 13 5
Round 7:
2 1 4 2 1 0 4 || 3 10 4 5 2 11 2
1 3 1 1 0 0 6 || 8 10 13 9 7 13 0
2 [1] 1 1 0 2 0 || 11 [15] 12 8 12 14 15
5 3 0 3 3 2 5 || 3 8 10 3 8 15 5
1 1 0 0 1 2 4 || 8 8 7 7 9 13 5
Round 8:
2 1 4 2 1 0 4 || 1 7 2 4 2 11 2
0 2 0 1 0 0 6 || 7 7 12 7 7 13 0
1 1 0 1 0 2 0 || 8 8 10 6 12 14 15
4 2 0 3 3 [2] 5 || 2 6 8 2 8 [15] 5
1 1 0 0 1 2 4 || 6 6 6 7 9 13 5
Round 9:
2 1 4 2 1 0 4 || 1 7 2 4 2 11 2
0 2 0 1 0 0 6 || 7 7 12 7 6 12 0
1 1 0 1 0 [1] 0 || 8 8 10 5 10 [13] 13
4 2 0 3 2 2 4 || 2 6 8 0 6 9 3
1 1 0 0 0 1 3 || 6 6 6 5 8 10 4
Round 10:
2 1 4 2 1 0 4 || 1 7 2 4 2 10 1
0 2 [0] 1 0 0 5 || 7 7 [12] 7 6 11 0
1 1 0 1 0 1 0 || 8 8 10 4 8 9 10
4 2 0 3 1 1 3 || 2 6 8 0 6 8 3
1 1 0 0 0 1 3 || 6 6 6 4 6 7 2
Round 11:
2 0 3 1 1 0 4 || 0 6 0 3 0 10 1
0 1 0 0 0 [0] 5 || 4 5 5 5 3 [11] 0
1 0 0 0 0 1 0 || 6 8 6 4 6 9 10
4 2 0 3 1 1 3 || 1 5 6 0 5 8 3
1 1 0 0 0 1 3 || 6 6 6 4 6 7 2
Round 12:
2 0 3 1 0 0 3 || 0 6 0 2 1 7 1
0 1 0 0 0 0 4 || 4 5 5 4 1 7 0
1 0 0 0 0 [0] 0 || 6 8 6 4 5 [9] 8
4 2 0 3 1 1 3 || 1 5 6 0 4 7 2
1 1 0 0 0 1 3 || 6 6 6 4 6 7 2
Round 13:
2 0 3 1 0 0 3 || 0 6 0 2 1 6 0
0 1 0 0 0 0 3 || 4 5 5 4 1 6 0
1 [0] 0 0 0 0 0 || 6 [8] 6 3 3 5 5
4 2 0 3 0 0 2 || 1 5 6 0 4 6 2
1 1 0 0 0 1 3 || 6 6 6 3 4 4 0
Round 14:
2 0 3 1 0 [0] 3 || 0 5 0 2 1 [6] 0
0 0 0 0 0 0 3 || 2 5 4 4 1 6 0
0 0 0 0 0 0 0 || 4 4 4 3 3 5 5
3 1 0 3 0 0 2 || 0 4 5 0 4 6 2
1 1 0 0 0 1 3 || 4 4 5 3 4 4 0
Round 15:
2 0 3 1 0 0 2 || 0 5 0 2 1 4 0
0 0 0 0 0 0 2 || 2 5 4 4 1 4 0
0 0 0 0 0 0 0 || 4 4 4 3 3 4 4
3 1 0 3 0 [0] 2 || 0 4 5 0 4 [6] 2
1 1 0 0 0 1 3 || 4 4 5 3 4 4 0
Round 16:
2 [0] 3 1 0 0 2 || 0 [5] 0 2 1 4 0
0 0 0 0 0 0 2 || 2 5 4 4 1 4 0
0 0 0 0 0 0 0 || 4 4 4 3 3 3 3
3 1 0 3 0 0 1 || 0 4 5 0 3 3 1
1 1 0 0 0 0 2 || 4 4 5 3 3 3 0
Round 17:
1 0 2 1 0 0 2 || 0 3 0 1 1 4 0
0 0 0 0 0 0 2 || 1 3 3 3 1 4 0
0 0 0 0 0 0 0 || 4 4 4 3 3 3 3
3 1 [0] 3 0 0 1 || 0 4 [5] 0 3 3 1
1 1 0 0 0 0 2 || 4 4 5 3 3 3 0
Round 18:
1 0 2 1 0 0 2 || 0 3 0 1 1 4 0
0 0 0 0 0 0 2 || 1 3 3 3 1 4 0
0 0 0 0 0 0 0 || 3 3 2 2 2 3 3
3 [0] 0 2 0 0 1 || 0 [4] 2 0 2 3 1
1 0 0 0 0 0 2 || 2 4 2 2 2 3 0
Round 19:
1 0 2 1 0 [0] 2 || 0 3 0 1 1 [4] 0
0 0 0 0 0 0 2 || 1 3 3 3 1 4 0
0 0 0 0 0 0 0 || 2 2 2 2 2 3 3
2 0 0 2 0 0 1 || 0 2 2 0 2 3 1
0 0 0 0 0 0 2 || 2 2 2 2 2 3 0
Round 20:
1 [0] 2 1 0 0 1 || 0 [3] 0 1 1 2 0
0 0 0 0 0 0 1 || 1 3 3 3 1 2 0
0 0 0 0 0 0 0 || 2 2 2 2 2 2 2
2 0 0 2 0 0 1 || 0 2 2 0 2 3 1
0 0 0 0 0 0 2 || 2 2 2 2 2 3 0
Round 21:
0 0 1 1 0 0 1 || 0 1 0 0 1 2 0
0 0 0 0 0 0 1 || 0 1 2 2 1 2 0
0 0 0 0 0 0 0 || 2 2 2 2 2 2 2
2 0 0 2 0 [0] 1 || 0 2 2 0 2 [3] 1
0 0 0 0 0 0 2 || 2 2 2 2 2 3 0
Round 22:
0 0 1 1 0 0 1 || 0 1 0 0 1 2 0
0 0 0 0 0 0 1 || 0 1 2 2 1 2 0
[0] 0 0 0 0 0 0 || [2] 2 2 2 2 1 1
2 0 0 2 0 0 0 || 0 2 2 0 2 1 1
0 0 0 0 0 0 1 || 2 2 2 2 2 1 0
Round 23:
0 0 1 1 0 0 1 || 0 1 0 0 1 2 0
0 0 [0] 0 0 0 1 || 0 1 [2] 2 1 2 0
0 0 0 0 0 0 0 || 1 1 2 2 2 1 1
1 0 0 2 0 0 0 || 0 1 2 0 2 1 1
0 0 0 0 0 0 1 || 1 1 2 2 2 1 0
Round 24:
0 0 0 0 0 0 1 || 0 0 0 0 0 2 0
0 0 0 0 0 0 1 || 0 0 0 0 0 2 0
0 0 [0] 0 0 0 0 || 1 1 [2] 2 2 1 1
1 0 0 2 0 0 0 || 0 1 2 0 2 1 1
0 0 0 0 0 0 1 || 1 1 2 2 2 1 0
Round 25:
0 0 0 0 0 [0] 1 || 0 0 0 0 0 [2] 0
0 0 0 0 0 0 1 || 0 0 0 0 0 2 0
0 0 0 0 0 0 0 || 1 1 1 1 1 1 1
1 0 0 1 0 0 0 || 0 1 1 0 1 1 1
0 0 0 0 0 0 1 || 1 1 1 1 1 1 0
Round 26:
0 0 0 0 0 0 0 || 0 0 0 0 0 0 0
0 0 0 0 0 0 0 || 0 0 0 0 0 0 0
[0] 0 0 0 0 0 0 || [1] 1 1 1 1 0 0
1 0 0 1 0 0 0 || 0 1 1 0 1 1 1
0 0 0 0 0 0 1 || 1 1 1 1 1 1 0
Round 27:
0 0 0 0 0 0 0 || 0 0 0 0 0 0 0
0 0 0 0 0 0 0 || 0 0 0 0 0 0 0
0 0 [0] 0 0 0 0 || 0 0 [1] 1 1 0 0
0 0 0 1 0 0 0 || 0 0 1 0 1 1 1
0 0 0 0 0 0 1 || 0 0 1 1 1 1 0
Round 28:
0 0 0 0 0 0 0 || 0 0 0 0 0 0 0
0 0 0 0 0 0 0 || 0 0 0 0 0 0 0
0 0 0 0 0 0 0 || 0 0 0 0 0 0 0
0 0 0 0 0 [0] 0 || 0 0 0 0 0 [1] 1
0 0 0 0 0 0 1 || 0 0 0 0 0 1 0
Done in 28 rounds
其他回答
使用分支和定界的数学整数线性规划
As it has already been mentioned, this problem can be solved using integer linear programming (which is NP-Hard). Mathematica already has ILP built in. "To solve an integer linear programming problem Mathematica first solves the equational constraints, reducing the problem to one containing inequality constraints only. Then it uses lattice reduction techniques to put the inequality system in a simpler form. Finally, it solves the simplified optimization problem using a branch-and-bound method." [see Constrained Optimization Tutorial in Mathematica.. ]
我写了下面的代码，利用ILP库的Mathematica。它的速度快得惊人。
solveMatrixBombProblem[problem_, r_, c_] :=
Module[{},
bombEffect[x_, y_, m_, n_] :=
Table[If[(i == x || i == x - 1 || i == x + 1) && (j == y ||
j == y - 1 || j == y + 1), 1, 0], {i, 1, m}, {j, 1, n}];
bombMatrix[m_, n_] :=
Transpose[
Table[Table[
Part[bombEffect[(i - Mod[i, n])/n + 1, Mod[i, n] + 1, m,
n], (j - Mod[j, n])/n + 1, Mod[j, n] + 1], {j, 0,
m*n - 1}], {i, 0, m*n - 1}]];
X := x /@ Range[c*r];
sol = Minimize[{Total[X],
And @@ Thread[bombMatrix[r, c].X >= problem] &&
And @@ Thread[X >= 0] && Total[X] <= 10^100 &&
Element[X, Integers]}, X];
Print["Minimum required bombs = ", sol[[1]]];
Print["A possible solution = ",
MatrixForm[
Table[x[c*i + j + 1] /. sol[[2]], {i, 0, r - 1}, {j, 0,
c - 1}]]];]
对于问题中提供的示例:
solveMatrixBombProblem[{2, 3, 4, 7, 1, 1, 5, 2, 6, 2, 4, 3, 4, 2, 1, 2, 1, 2, 4, 1, 3, 1, 3, 4, 1, 2, 1, 4, 3, 2, 6, 9, 1, 6, 4}, 7, 5]
输出
对于那些用贪婪算法读这篇文章的人
在下面这个10x10的问题上试试你的代码:
5 20 7 1 9 8 19 16 11 3
17 8 15 17 12 4 5 16 8 18
4 19 12 11 9 7 4 15 14 6
17 20 4 9 19 8 17 2 10 8
3 9 10 13 8 9 12 12 6 18
16 16 2 10 7 12 17 11 4 15
11 1 15 1 5 11 3 12 8 3
7 11 16 19 17 11 20 2 5 19
5 18 2 17 7 14 19 11 1 6
13 20 8 4 15 10 19 5 11 12
这里用逗号分隔:
5, 20, 7, 1, 9, 8, 19, 16, 11, 3, 17, 8, 15, 17, 12, 4, 5, 16, 8, 18, 4, 19, 12, 11, 9, 7, 4, 15, 14, 6, 17, 20, 4, 9, 19, 8, 17, 2, 10, 8, 3, 9, 10, 13, 8, 9, 12, 12, 6, 18, 16, 16, 2, 10, 7, 12, 17, 11, 4, 15, 11, 1, 15, 1, 5, 11, 3, 12, 8, 3, 7, 11, 16, 19, 17, 11, 20, 2, 5, 19, 5, 18, 2, 17, 7, 14, 19, 11, 1, 6, 13, 20, 8, 4, 15, 10, 19, 5, 11, 12
对于这个问题，我的解决方案包含208个炸弹。这里有一个可能的解决方案(我能够在大约12秒内解决这个问题)。
作为一种测试Mathematica产生结果的方法，看看你的贪婪算法是否能做得更好。
蛮力!
我知道它效率不高，但即使你找到了一个更快的算法，你也可以对这个结果进行测试，以了解它有多准确。
使用一些递归，像这样:
void fn(tableState ts, currentlevel cl)
{
// first check if ts is all zeros yet, if not:
//
// do a for loop to go through all cells of ts,
// for each cell do a bomb, and then
// call:
// fn(ts, cl + 1);
}
你可以通过缓存来提高效率，如果不同的方法导致相同的结果，你不应该重复相同的步骤。
阐述:
如果轰炸单元格1,3,5的结果与轰炸单元格5,3,1的结果相同，那么，对于这两种情况，您不应该重新执行所有后续步骤，只需1就足够了，您应该将所有表状态存储在某个地方并使用其结果。
表统计信息的散列可以用于快速比较。
这是对第一个问题的回答。我没有注意到他改变了参数。
创建一个所有目标的列表。根据掉落物品(掉落物品本身和所有邻居)影响的正数值的数量为目标分配一个值。最高值是9。
根据受影响目标的数量(降序)对目标进行排序，对每个受影响目标的和进行二次降序排序。
向排名最高的目标投掷炸弹，然后重新计算目标，直到所有目标值都为零。
同意，这并不总是最优的。例如,
100011
011100
011100
011100
000000
100011
这种方法需要5枚炸弹才能清除。最理想的情况是，你可以在4分钟内完成。不过,很 非常接近，没有回头路。在大多数情况下，这将是最优的，或者非常接近。
使用原来的问题数，该方法解决28个炸弹。
添加代码来演示这种方法(使用带有按钮的表单):
private void button1_Click(object sender, EventArgs e)
{
int[,] matrix = new int[10, 10] {{5, 20, 7, 1, 9, 8, 19, 16, 11, 3},
{17, 8, 15, 17, 12, 4, 5, 16, 8, 18},
{ 4, 19, 12, 11, 9, 7, 4, 15, 14, 6},
{ 17, 20, 4, 9, 19, 8, 17, 2, 10, 8},
{ 3, 9, 10, 13, 8, 9, 12, 12, 6, 18},
{16, 16, 2, 10, 7, 12, 17, 11, 4, 15},
{ 11, 1, 15, 1, 5, 11, 3, 12, 8, 3},
{ 7, 11, 16, 19, 17, 11, 20, 2, 5, 19},
{ 5, 18, 2, 17, 7, 14, 19, 11, 1, 6},
{ 13, 20, 8, 4, 15, 10, 19, 5, 11, 12}};
int value = 0;
List<Target> Targets = GetTargets(matrix);
while (Targets.Count > 0)
{
BombTarget(ref matrix, Targets[0]);
value += 1;
Targets = GetTargets(matrix);
}
Console.WriteLine( value);
MessageBox.Show("done: " + value);
}
private static void BombTarget(ref int[,] matrix, Target t)
{
for (int a = t.x - 1; a <= t.x + 1; a++)
{
for (int b = t.y - 1; b <= t.y + 1; b++)
{
if (a >= 0 && a <= matrix.GetUpperBound(0))
{
if (b >= 0 && b <= matrix.GetUpperBound(1))
{
if (matrix[a, b] > 0)
{
matrix[a, b] -= 1;
}
}
}
}
}
Console.WriteLine("Dropped bomb on " + t.x + "," + t.y);
}
private static List<Target> GetTargets(int[,] matrix)
{
List<Target> Targets = new List<Target>();
int width = matrix.GetUpperBound(0);
int height = matrix.GetUpperBound(1);
for (int x = 0; x <= width; x++)
{
for (int y = 0; y <= height; y++)
{
Target t = new Target();
t.x = x;
t.y = y;
SetTargetValue(matrix, ref t);
if (t.value > 0) Targets.Add(t);
}
}
Targets = Targets.OrderByDescending(x => x.value).ThenByDescending( x => x.sum).ToList();
return Targets;
}
private static void SetTargetValue(int[,] matrix, ref Target t)
{
for (int a = t.x - 1; a <= t.x + 1; a++)
{
for (int b = t.y - 1; b <= t.y + 1; b++)
{
if (a >= 0 && a <= matrix.GetUpperBound(0))
{
if (b >= 0 && b <= matrix.GetUpperBound(1))
{
if (matrix[ a, b] > 0)
{
t.value += 1;
t.sum += matrix[a,b];
}
}
}
}
}
}
你需要的一个类:
class Target
{
public int value;
public int sum;
public int x;
public int y;
}
这里有一个解决方案，推广良好的性质的角。
让我们假设我们可以为给定的字段找到一个完美的落点，也就是说，一个减少其中值的最佳方法。然后，为了找到最少的炸弹数量，一个算法的初稿可能是(代码是从ruby实现中复制粘贴的):
dropped_bomb_count = 0
while there_are_cells_with_non_zero_count_left
coordinates = choose_a_perfect_drop_point
drop_bomb(coordinates)
dropped_bomb_count += 1
end
return dropped_bomb_count
挑战是choose_a_perfect_drop_point。首先，让我们定义一个完美的落点是什么。
(x, y)的放置点会减少(x, y)中的值。它也可能会减少其他单元格中的值。 (x, y)的放置点A比(x, y)的放置点b更好，如果它减少了b所减少的单元格的适当超集中的值。 如果没有其他更好的投放点，投放点是最大的。 (x, y)的两个放置点是等效的，如果它们减少了同一组单元格。 如果(x, y)的放置点等价于(x, y)的所有最大放置点，那么它就是完美的。
如果(x, y)存在一个完美的投放点，那么您不能比在(x, y)的一个完美投放点上投放炸弹更有效地降低(x, y)处的值。
给定字段的完美放置点是其任何单元格的完美放置点。
以下是一些例子:
1 0 1 0 0
0 0 0 0 0
1 0 0 0 0
0 0 0 0 0
0 0 0 0 0
单元格(0,0)(从零开始的索引)的完美放置点是(1,1)。(1,1)的所有其他放置点，即(0,0)、(0,1)和(1,0)，减少的单元格较少。
0 0 0 0 0
0 0 0 0 0
0 0 1 0 0
0 0 0 0 0
0 0 0 0 0
单元格(2,2)(从零开始的索引)的完美落点是(2,2)，以及所有周围的单元格(1,1)、(1,2)、(1,3)、(2,1)、(2,3)、(3,1)、(3,2)和(3,3)。
0 0 0 0 1
0 0 0 0 0
0 0 1 0 0
0 0 0 0 0
0 0 0 0 0
单元格(2,2)的完美放置点是(3,1):它减少了(2,2)中的值和(4,0)中的值。(2,2)的所有其他放置点都不是最大的，因为它们减少了一个单元格。(2,2)的完美下拉点也是(4,0)的完美下拉点，它是字段的唯一完美下拉点。它为这个领域带来了完美的解决方案(一颗炸弹)。
1 0 0 0 0
0 0 0 0 0
0 0 1 0 0
0 0 0 0 0
1 0 0 0 0
(2,2)没有完美的落点:(1,1)和(1,3)都减少(2,2)和另一个单元格(它们是(2,2)的最大落点)，但它们不相等。然而，(1,1)是(0,0)的完美落点，(1,3)是(0,4)的完美落点。
根据完美落点的定义和一定的检查顺序，我得到了以下问题示例的结果:
Drop bomb on 1, 1
Drop bomb on 1, 1
Drop bomb on 1, 5
Drop bomb on 1, 5
Drop bomb on 1, 5
Drop bomb on 1, 6
Drop bomb on 1, 2
Drop bomb on 1, 2
Drop bomb on 0, 6
Drop bomb on 0, 6
Drop bomb on 2, 1
Drop bomb on 2, 5
Drop bomb on 2, 5
Drop bomb on 2, 5
Drop bomb on 3, 1
Drop bomb on 3, 0
Drop bomb on 3, 0
Drop bomb on 3, 0
Drop bomb on 3, 0
Drop bomb on 3, 0
Drop bomb on 3, 4
Drop bomb on 3, 4
Drop bomb on 3, 3
Drop bomb on 3, 3
Drop bomb on 3, 6
Drop bomb on 3, 6
Drop bomb on 3, 6
Drop bomb on 4, 6
28
然而，该算法只有在每一步之后至少有一个完美落点时才能工作。可以在没有完美落点的情况下构建例子:
0 1 1 0
1 0 0 1
1 0 0 1
0 1 1 0
对于这些情况，我们可以修改算法，这样我们就不会选择完美的落点，而是选择一个具有最大落点的最小选择的坐标，然后计算每个选择的最小值。在上面的例子中，所有有值的单元格都有两个最大落点。例如，(0,1)有最大落点(1,1)和(1,2)。选择其中任何一个，然后计算最小值，会得到这样的结果:
Drop bomb on 1, 1
Drop bomb on 2, 2
Drop bomb on 1, 2
Drop bomb on 2, 1
2
你的新问题，有跨行不递减的值，很容易解决。
Observe that the left column contains the highest numbers. Therefore, any optimal solution must first reduce this column to zero. Thus, we can perform a 1-D bombing run over this column, reducing every element in it to zero. We let the bombs fall on the second column so they do maximum damage. There are many posts here dealing with the 1D case, I think, so I feel safe in skipping that case. (If you want me to describe it, I can.). Because of the decreasing property, the three leftmost columns will all be reduced to zero. But, we will provably use a minimum number of bombs here because the left column must be zeroed.
现在，一旦左边的列归零，我们只要剪掉最左边的三列现在归零，然后对现在化简的矩阵重复这一步骤。这必须给我们一个最优的解决方案，因为在每个阶段我们使用可证明的最少数量的炸弹。
