Maximize array sum after K negations using Priority Queue

Given an array of size n and a number k. We must modify array K number of times. Here modify array means in each operation we can replace any array element arr[i] by -arr[i]. We need to perform this operation in such a way that after K operations, sum of array must be maximum?

Examples: 

Input : arr[] = {-2, 0, 5, -1, 2} 
        K = 4
Output: 10
// Replace (-2) by -(-2), array becomes {2, 0, 5, -1, 2}
// Replace (-1) by -(-1), array becomes {2, 0, 5, 1, 2}
// Replace (0) by -(0), array becomes {2, 0, 5, 1, 2}
// Replace (0) by -(0), array becomes {2, 0, 5, 1, 2}

Input : arr[] = {9, 8, 8, 5} 
        K = 3
Output: 20

We strongly recommend that you click here and practice it, before moving on to the solution.

We have discussed a O(nk) solution in below post.
Maximize array sum after K negations | Set 1

The idea used in above post is to replace the minimum element arr[i] in array by -arr[i] for current operation. In this way we can make sum of array maximum after K operations. Once interesting case is, once minimum element becomes 0, we don’t need to make any more changes.

The implementation used in above solution uses linear search to find minimum element. The time complexity of the above discussed solution is O(nk)
In this post an optimized solution is implemented that uses a priority queue (or binary heap) to find minimum element quickly. 

Below is the implementation of the idea. It uses PriorityQueue class in Java.

C++

// A PriorityQueue based C++ program to
// maximize array sum after k negations.
#include <bits/stdc++.h>
using namespace std;

// Function to find Maximum sum
// after K negations
int MaxSum(int a[], int n, int k)
{
    int sum = 0;

    // Create a min heap for priority queue
    priority_queue<int, vector<int>, greater<int>> pq;

    // Insert all elements in f array in priority_queue
    for(int i = 0; i < n; i++)
    {
        pq.push(a[i]);
    }

    while (k--)
    {

        // Retrieve and remove min element
        int temp = pq.top();

        pq.pop();

        // Modify the minimum element and
        // add back to priority queue
        temp = (temp) * -1;
        pq.push(temp);
    }

    // Calculate the sum
    while (!pq.empty())
    {
        sum = sum + pq.top();
        pq.pop();
    }
    return sum;
}

// Driver Code
int main()
{
    int a[] = { -2, 0, 5, -1, 2 };
    int n = sizeof(a) / sizeof(a[0]);
    int k = 4;

    cout << MaxSum(a, n, k);
    return 0;
}

// This code is contributed by Harshit Srivastava

Java

// A PriorityQueue based Java program to maximize array
// sum after k negations.
import java.util.*;

class maximizeSum
{
    public static int maxSum(int[] a, int k)
    {
        // Create a priority queue and insert all array elements
        // int
        PriorityQueue<Integer> pq = new PriorityQueue<>();
        for (int x : a)
            pq.add(x);

        // Do k negations by removing a minimum element k times
        while (k-- > 0)
        {
            // Retrieve and remove min element
            int temp = pq.poll();

            // Modify the minimum element and add back
            // to priority queue
            temp *= -1;
            pq.add(temp);
        }

        // Compute sum of all elements in priority queue.
        int sum = 0;
        for (int x : pq)
            sum += x;
        return sum;
    }

    // Driver code
    public static void main (String[] args)
    {
        int[] arr = {-2, 0, 5, -1, 2};
        int k = 4;
        System.out.println(maxSum(arr, k));
    }
}

Python3

# Python code
# A PriorityQueue based Python program to
# maximize array sum after k negations.

# import library
import heapq

# Function to find Maximum sum
# after K negations
def MaxSum(a, n, k):
    sum = 0

    # Create a min heap for priority queue
    pq = []

    # Insert all elements in f array in priority_queue
    for i in range(n):
        heapq.heappush(pq, a[i])

    while k > 0:

        # Retrieve and remove min element
        temp = heapq.heappop(pq)

        # Modify the minimum element and
        # add back to priority queue
        temp = (temp) * -1
        heapq.heappush(pq, temp)
        k -= 1

    # Calculate the sum
    while (pq):
        sum += heapq.heappop(pq)

    return sum

# Driver Code
a = [-2, 0, 5, -1, 2]
n = len(a)
k = 4

print(MaxSum(a, n, k))

# This code is contributed by akashish__

C#

// A PriorityQueue based C# program to maximize array
// sum after k negations.
using System;
using System.Collections.Generic;

public class GFG {

    public static int maxSum(int[] a, int k)
    {
        // Create a priority queue and insert all array
        // elements int
        List<int> pq = new List<int>(a);

        // Do k negations by removing a minimum element
        // k times
        while (k-- > 0) {
            // Retrieve and remove min element
            pq.Sort();
            int temp = pq[0];
            pq.Remove(pq[0]);

            // Modify the minimum element and add back
            // to priority queue
            temp *= -1;
            pq.Add(temp);
        }

        // Compute sum of all elements in priority queue.
        int sum = 0;
        foreach(int x in pq) sum += x;
        return sum;
    }

    // Driver code
    static public void Main()
    {
        int[] arr = { -2, 0, 5, -1, 2 };
        int k = 4;
        Console.WriteLine(maxSum(arr, k));
    }
}
// contributed by akashish__

Javascript

function maxSum(a, k)
    {
        let pq = [];
        for (let i=0;i<a.length;i++)
            pq.push(a[i]);

        pq.sort();

        // Do k negations by removing a minimum element k times
        while (k-- > 0)
        {
            // Retrieve and remove min element
            let temp = pq.shift();

            // Modify the minimum element and add back
            // to priority queue
            temp *= -1;
            pq.push(temp);
            pq.sort();
        }

        // Compute sum of all elements in priority queue.
        let sum = 0;
        for (let i=0;i<pq.length;i++)
            sum += pq[i];

        return sum;
    }


        let arr = [-2, 0, 5, -1, 2];
        let k = 4;
        console.log(maxSum(arr, k));

// This code is contributed by aadityaburujwale.

Output:

10

Time Complexity: O(N log N), where N is the size of array A[].
Auxiliary Space: O(N) 

This article is contributed by Prakhar. If you like GeeksforGeeks and would like to contribute, you can also write an article using write.geeksforgeeks.org or mail your article to review-team@geeksforgeeks.org. See your article appearing on the GeeksforGeeks main page and help other Geeks.
Please write comments if you find anything incorrect, or you want to share more information about the topic discussed above.