Minimize cost of swapping set bits with unset bits in a given Binary string

Given a binary string S of size N, the task is to find the minimum cost by swapping every set bit with an unset bit such that the cost of swapping pairs of bits at indices i and j is abs(j – i).

Note: A swapped bit can’t be swapped twice and the count of set bit in the given binary string is at most N/2.

Examples:

Input: S = “1010001”
Output: 3
Explanation:
Following the swapping of characters required:

1. Swap characters at indices (0, 1) modifies the string to “0110001” and the cost of this operation is |1 – 0| = 1.
2. Swap characters at indices (2, 3) modifies the string to “0101001” and the cost of this operation is |2 – 1| = 1.
3. Swap characters at indices (6, 7) modifies the string to “0101010” and the cost of this operation is |7 – 6| = 1.

After the above operations, all the set bits is replaced with unset bits and the total cost of operations is 1 + 1 + 1 = 3.

Input: S = “1100”
Output: 4

Approach: The given problem can be solved using Dynamic Programming by storing the indices of set and unset bits in two auxiliary arrays, say A[] and B[], and then find the sum of the difference between array elements A[] with any element of the array B[]. Follow the steps below to solve the given problem:

• Initialize two arrays, say A[] and B[], and store the indices of set and unset bits in it.
• Initialize a 2D array, dp[][] of dimensions K*(N – K) where K is the count of set bit in S such thatdp[i][j] stores the minimum cost of swapping the i^th array element A[] with the j^th array element B[].
• Now, for each state there are two choices:
  1. Swap the i^th array element A[] till the (j – 1)^th array element B[] as dp[i][j] = dp[i][j – 1].
  2. Swap the (i – 1)^th array element A[] till the (j – 1)^th array element B[] and the i^th array element A[] with j^th array element B[] and this state can be calculated as dp[i][j] = dp[i – 1][j – 1] + abs(A[i] – B[i]).
• Now, choose the minimum of the above two choices to find the current state as:

 dp[i][j] = min(dp[i][j-1], dp[i-1][j-1] + abs(A[i] – B[j]))

• After completing the above steps, print the value of dp[K][N – K] as the resultant minimum number of operations.

Below is the implementation of the above approach:

C++

// C++ program for the above approach

#include <bits/stdc++.h>
using namespace std;
#define INF 1000000000

// Function to find the minimum cost
// required to swap every set bit with
// an unset bit
int minimumCost(string s)
{
    int N = s.length();

    // Stores the indices of set and
    // unset bits of the string S
    vector<int> A, B;

    // Traverse the string S
    for (int i = 0; i < N; i++) {

        // Store the indices
        if (s[i] == '1') {
            A.push_back(i);
        }
        else {
            B.push_back(i);
        }
    }

    int n1 = A.size();
    int n2 = B.size();

    // Initialize a dp table of size
    // n1*n2
    int dp[n1 + 1][n2 + 1];

    // Initialize all states to 0
    memset(dp, 0, sizeof(dp));

    // Set unreachable states to INF
    for (int i = 1; i <= n1; i++) {
        dp[i][0] = INF;
    }

    // Fill the dp Table according to
    // the given recurrence relation
    for (int i = 1; i <= n1; i++) {
        for (int j = 1; j <= n2; j++) {

            // Update the value of
            // dp[i][j]
            dp[i][j] = min(
                dp[i][j - 1],
                dp[i - 1][j - 1]
                    + abs(A[i - 1] - B[j - 1]));
        }
    }

    // Return the minimum cost
    return dp[n1][n2];
}

// Driver Code
int main()
{
    string S = "1010001";
    cout << minimumCost(S);

    return 0;
}

Java

// Java program for the above approach
import java.util.*;

class GFG{
static final int INF = 1000000000;

// Function to find the minimum cost
// required to swap every set bit with
// an unset bit
static int minimumCost(String s)
{
    int N = s.length();

    // Stores the indices of set and
    // unset bits of the String S
    Vector<Integer> A = new Vector<Integer>();
    Vector<Integer> B = new Vector<Integer>();

    // Traverse the String S
    for (int i = 0; i < N; i++) {

        // Store the indices
        if (s.charAt(i) == '1') {
            A.add(i);
        }
        else {
            B.add(i);
        }
    }

    int n1 = A.size();
    int n2 = B.size();

    // Initialize a dp table of size
    // n1*n2
    int [][]dp = new int[n1 + 1][n2 + 1];


    // Set unreachable states to INF
    for (int i = 1; i <= n1; i++) {
        dp[i][0] = INF;
    }

    // Fill the dp Table according to
    // the given recurrence relation
    for (int i = 1; i <= n1; i++) {
        for (int j = 1; j <= n2; j++) {

            // Update the value of
            // dp[i][j]
            dp[i][j] = Math.min(
                dp[i][j - 1],
                dp[i - 1][j - 1]
                    + Math.abs(A.get(i - 1) - B.get(j - 1)));
        }
    }

    // Return the minimum cost
    return dp[n1][n2];
}

// Driver Code
public static void main(String[] args)
{
    String S = "1010001";
    System.out.print(minimumCost(S));
}
}

// This code is contributed by shikhasingrajput

Python3

# Python program for the above approach
INF = 1000000000

# Function to find the minimum cost
# required to swap every set bit with
# an unset bit


def minimumCost(s):
    N = len(s)

    # Stores the indices of set and
    # unset bits of the string S
    A = []
    B = []

    # Traverse the string S
    for i in range(0, N):

        # Store the indices
        if (s[i] == "1"):
            A.append(i)
        else:
            B.append(i)

    n1 = len(A)
    n2 = len(B)

    # Initialize a dp table of size
    # n1*n2
    dp = [[0 for i in range(n2 + 1)] for j in range(n1 + 1)]

    # Set unreachable states to INF
    for i in range(1, n1 + 1):
        dp[i][0] = INF

    # Fill the dp Table according to
    # the given recurrence relation
    for i in range(1, n1 + 1):
        for j in range(1, n2 + 1):

            # Update the value of
            # dp[i][j]
            dp[i][j] = min(
                dp[i][j - 1],
                dp[i - 1][j - 1] + abs(A[i - 1] - B[j - 1])
            )

    # Return the minimum cost
    return dp[n1][n2]


# Driver Code
S = "1010001"
print(minimumCost(S))

# This code is contributed by _saurabh_jaiswal.

C#

// C# program for the above approach
using System;
using System.Collections;
using System.Collections.Generic;

public class Program
{

// Function to find the minimum cost
// required to swap every set bit with
// an unset bit
static int minimumCost(string s)
{
    int INF = 1000000000;
    int N = s.Length;

    // Stores the indices of set and
    // unset bits of the string S
    List<int> A = new List<int>();
    List<int> B = new List<int>();

    // Traverse the string S
    for (int i = 0; i < N; i++) {

        // Store the indices
        if (s[i] == '1') {
            A.Add(i);
        }
        else {
            B.Add(i);
        }
    }

    int n1 = A.Count;
    int n2 = B.Count;

    // Initialize a dp table of size
    // n1*n2
    int [,]dp = new  int[n1 + 1,n2 + 1];


    // Set unreachable states to INF
    for (int i = 1; i <= n1; i++) {
        dp[i,0] = INF;
    }

    // Fill the dp Table according to
    // the given recurrence relation
    for (int i = 1; i <= n1; i++) {
        for (int j = 1; j <= n2; j++) {

            // Update the value of
            // dp[i][j]
            dp[i,j] = Math.Min(
                dp[i,j - 1],
                dp[i - 1,j - 1]
                    + Math.Abs(A[i - 1] - B[j - 1]));
        }
    }

    // Return the minimum cost
    return dp[n1,n2];
}

    public static void Main()
    {
        string S = "1010001";
        Console.Write(minimumCost(S));
    }
}

// This code is contributed by rutvik_56.

Javascript

<script>
// Javascript program for the above approach

let INF = 1000000000;

// Function to find the minimum cost
// required to swap every set bit with
// an unset bit
function minimumCost(s) {
  let N = s.length;

  // Stores the indices of set and
  // unset bits of the string S
  let A = [],
    B = [];

  // Traverse the string S
  for (let i = 0; i < N; i++) {
    // Store the indices
    if (s[i] == "1") {
      A.push(i);
    } else {
      B.push(i);
    }
  }

  let n1 = A.length;
  let n2 = B.length;

  // Initialize a dp table of size
  // n1*n2
  let dp = new Array(n1 + 1).fill(0).map(() => new Array(n2 + 1).fill(0));

  // Set unreachable states to INF
  for (let i = 1; i <= n1; i++) {
    dp[i][0] = INF;
  }

  // Fill the dp Table according to
  // the given recurrence relation
  for (let i = 1; i <= n1; i++) {
    for (let j = 1; j <= n2; j++) {
      // Update the value of
      // dp[i][j]
      dp[i][j] = Math.min(
        dp[i][j - 1],
        dp[i - 1][j - 1] + Math.abs(A[i - 1] - B[j - 1])
      );
    }
  }

  // Return the minimum cost
  return dp[n1][n2];
}

// Driver Code

let S = "1010001";
document.write(minimumCost(S));

// This code is contributed by gfgking.
</script>

Output

3

Time Complexity: O(K*(N – K)) where K is the count of set bit in S.
Auxiliary Space: O(K*(N – K))

Efficient approach : Space optimization

In previous approach the current value dp[i][j] is only depend upon the current and previous row values of DP. So to optimize the space complexity we use a single 1D array to store the computations.

Implementation steps:

• Create a 1D vector dp of size n+1 and initialize it with 0.
• Set a base case by initializing the values of DP .
• Now iterate over subproblems by the help of nested loop and get the current value from previous computations.
• At last return and print the final answer stored in dp[n].

Implementation: 

C++

#include <bits/stdc++.h>
using namespace std;
#define INF 1000000000

// Function to find the minimum cost
// required to swap every set bit with
// an unset bit
int minimumCost(string s)
{
    int n1 = 0, n2 = 0;
    for (char c : s) {
        if (c == '1') n1++;
        else n2++;
    }

    // Initialize a dp table of size
    // n1*n2
    int dp[n1 + 1];

    // Initialize all states to 0
    memset(dp, 0, sizeof(dp));

    // Set unreachable states to INF
    for (int i = 1; i <= n1; i++) {
        dp[i] = INF;
    }

    // Fill the dp Table according to
    // the given recurrence relation
    for (char c : s) {
        for (int i = n1; i >= 1; i--) {
            if (c == '0') dp[i] = min(dp[i], dp[i - 1] + n2 - i + 1);
            else dp[i] = min(dp[i], dp[i - 1] + i - 1);
        }
    }

    // Return the minimum cost
    return dp[n1];
}

// Driver Code
int main()
{
    string S = "1010001";
    cout << minimumCost(S);

    return 0;
}

Java

import java.util.Arrays;

public class MinimumCost {

  static int INF = 1000000000;

  // Function to find the minimum cost
  // required to swap every set bit with
  // an unset bit
  static int minimumCost(String s) {
    int n1 = 0, n2 = 0;
    for (char c : s.toCharArray()) {
      if (c == '1') {
        n1++;
      } else {
        n2++;
      }
    }

    // Initialize a dp table of size
    // n1*n2
    int[] dp = new int[n1 + 1];

    // Initialize all states to 0
    Arrays.fill(dp, 0);

    // Set unreachable states to INF
    for (int i = 1; i <= n1; i++) {
      dp[i] = INF;
    }

    // Fill the dp Table according to
    // the given recurrence relation
    for (char c : s.toCharArray()) {
      for (int i = n1; i >= 1; i--) {
        if (c == '0') {
          dp[i] = Math.min(dp[i], dp[i - 1] + n2 - i + 1);
        } else {
          dp[i] = Math.min(dp[i], dp[i - 1] + i - 1);
        }
      }
    }

    // Return the minimum cost
    return dp[n1];
  }

  // Driver Code
  public static void main(String[] args) {
    String S = "1010001";
    System.out.println(minimumCost(S));
  }
}

Python

INF = 1000000000

# Function to find the minimum cost
# required to swap every set bit with an unset bit
def minimumCost(s):
    n1 = s.count('1')
    n2 = len(s) - n1

    # Initialize a dp table of size n1*n2
    dp = [0]*(n1 + 1)

    # Set unreachable states to INF
    for i in range(1, n1 + 1):
        dp[i] = INF

    # Fill the dp Table according
    # to the given recurrence relation
    for c in s:
        for i in range(n1, 0, -1):
            if c == '0':
                dp[i] = min(dp[i], dp[i - 1] + n2 - i + 1)
            else:
                dp[i] = min(dp[i], dp[i - 1] + i - 1)

    # Return the minimum cost
    return dp[n1]

# Driver Code
if __name__ == "__main__":
    S = "1010001"
    print(minimumCost(S))

C#

using System;

namespace MinimumCostToSwapBits
{
    class Program
    {
        static int INF = 1000000000;

        static int MinimumCost(string s)
        {
            int n1 = 0, n2 = 0;
            foreach (char c in s)
            {
                if (c == '1') n1++;
                else n2++;
            }

            int[] dp = new int[n1 + 1];
            Array.Fill(dp, 0);

            for (int i = 1; i <= n1; i++)
            {
                dp[i] = INF;
            }

            foreach (char c in s)
            {
                for (int i = n1; i >= 1; i--)
                {
                    if (c == '0') dp[i] = Math.Min(dp[i], dp[i - 1] + n2 - i + 1);
                    else dp[i] = Math.Min(dp[i], dp[i - 1] + i - 1);
                }
            }

            return dp[n1];
        }

        static void Main(string[] args)
        {
            string S = "1010001";
            Console.WriteLine(MinimumCost(S));
        }
    }
}

Javascript

function minimumCost(s) {
  const INF = 1000000000;

  let n1 = 0, n2 = 0;
  for (let i = 0; i < s.length; i++) {
    if (s.charAt(i) == '1') n1++;
    else n2++;
  }

  const dp = new Array(n1 + 1).fill(0);

  for (let i = 1; i <= n1; i++) {
    dp[i] = INF;
  }

  for (let i = 0; i < s.length; i++) {
    const c = s.charAt(i);
    for (let j = n1; j >= 1; j--) {
      if (c == '0') dp[j] = Math.min(dp[j], dp[j - 1] + n2 - j + 1);
      else dp[j] = Math.min(dp[j], dp[j - 1] + j - 1);
    }
  }

  return dp[n1];
}

const S = "1010001";
console.log(minimumCost(S)); // Output: 3

Output

3

Time Complexity: O(N^2) 
Auxiliary Space: O(N)