Sum of absolute difference of all pairs raised to power K

Given an array arr[] of N integers and a number K, the task is to find the sum of absolute difference of all pairs raised to the power K in a given array i.e., .
Examples:

Input: arr[] = {1, 2, 3}, K = 1 
Output: 8 
Explanation:
Sum of |1-1|+|1-2|+|1-3|+|2-1|+|2-2|+|2-3|+|3-1|+|3-2|+|3-3| = 8 
Input: arr[] = {1, 2, 3}, K = 3 
Output: 20 
Explanation:
Sum of |1 – 1| ³ + |1 – 2| ³ + |1 – 3| ³ + |2 – 1| ³ + |2 – 2| ³ + |2 – 3| ³ + |3 – 1| ³ + |3 – 2| ³ + |3 – 3| ³ = 20 

Naive Approach: The idea is to generate all the possible pairs and find the absolute difference of each pair raised to the power K and sum up them together.
Time Complexity: O((log K)*N²)
Auxiliary Space: O(1)
Efficient Approach: We can improve the time complexity of the naive approach with the below calculations: 
For all possible pair, we have to find the value of 

Since for pairs (arr[i], arr[j]) the value of is being calculated twice. So the above equation can also be written as: 

Writing in terms of Binomial Equation: 

(Equation 1) 

Let Pre[i][a] = (Equation 2)
From Equation 1 and Equation 2, we have 

The value of Pre[i][a] can be calculated as: 

Pre[i][a] = {(-arr[1])^{K – a} + (-arr[2])^{K – a} … . . +(-arr[i – 1])^{K – a}}. 
So, Pre[i+1][a] = Pre[i][a]+(-arr[i])^{K – a}

Below is the implementation of the above approach: 

C++

// C++ program for the above approach

#include <bits/stdc++.h>
#define ll long long
using namespace std;

class Solution {

public:
    // Since K can be 100 max
    ll ncr[101][101];
    int n, k;
    vector<ll> A;

    // Constructor
    Solution(int N, int K, vector<ll> B)
    {
        // Initializing with -1
        memset(ncr, -1, sizeof(ncr));

        n = N;
        k = K;
        A = B;

        // Making vector A as 1-Indexing
        A.insert(A.begin(), 0);
    }

    ll f(int N, int K);

    ll pairsPower();
};

// To Calculate the value nCk
ll Solution::f(int n, int k)
{
    if (k == 0)
        return 1LL;

    if (n == k)
        return 1LL;

    if (n < k)
        return 0;

    if (ncr[n][k] != -1)
        return ncr[n][k];

    // Since nCj = (n-1)Cj + (n-1)C(j-1);
    return ncr[n][k] = f(n - 1, k)
                       + f(n - 1, k - 1);
}

// Function that summation of absolute
// differences of all pairs raised
// to the power k
ll Solution::pairsPower()
{
    ll pre[n + 1][k + 1];
    ll ans = 0;

    // Sort the given array
    sort(A.begin() + 1, A.end());

    // Precomputation part, O(n*k)
    for (int i = 1; i <= n; ++i) {
        pre[i][0] = 1LL;

        for (int j = 1; j <= k; j++) {
            pre[i][j] = A[i]
                        * pre[i][j - 1];
        }

        if (i != 1) {
            for (int j = 0; j <= k; ++j)
                pre[i][j] = pre[i][j]
                            + pre[i - 1][j];
        }
    }

    // Traverse the array arr[]
    for (int i = n; i >= 2; --i) {

        // For each K
        for (int j = 0; j <= k; j++) {

            ll val = f(k, j);

            ll val1 = pow(A[i], k - j)
                      * pre[i - 1][j];

            val = val * val1;

            if (j % 2 == 0)
                ans = (ans + val);

            else
                ans = (ans - val);
        }
    }

    ans = 2LL * ans;

    // Return the final answer
    return ans;
}

// Driver Code
int main()
{
    // Given N and K
    int N = 3;
    int K = 3;

    // Given array
    vector<ll> arr = { 1, 2, 3 };

    // Creation of Object of class
    Solution obj(N, K, arr);

    // Function Call
    cout << obj.pairsPower() << endl;
    return 0;
}

Java

/*package whatever //do not write package name here */
import java.util.*;

public class GFG{
  // Since K can be 100 max
  static long[][] ncr = new long[101][];
  static int n, k;
  static long A[];

  // To Calculate the value nCk
  static long f(int n, int k)
  {
    if (k == 0)
      return (long)1;

    if (n == k)
      return (long)1;

    if (n < k)
      return 0;

    if (ncr[n][k] != -1)
      return ncr[n][k];

    ncr[n][k] = f(n - 1, k) + f(n - 1, k - 1);
    // Since nCj = (n-1)Cj + (n-1)C(j-1);
    return ncr[n][k];
  }

  // Function that summation of absolute
  // differences of all pairs raised
  // to the power k
  static long pairsPower()
  {
    long[][] pre = new long[n + 1][];
    long ans = 0;
    for(int i = 0 ; i <= n ; i++){
      pre[i] = new long[k + 1];
    }

    // Sort the given array
    Arrays.sort(A,1,A.length);

    // Precomputation part, O(n*k)
    for (int i = 1 ; i <= n ; ++i) {
      pre[i][0] = (long)1;

      for (int j = 1 ; j <= k ; j++) {
        pre[i][j] = A[i] * pre[i][j - 1];
      }

      if (i != 1) {
        for (int j = 0 ; j <= k ; ++j){
          pre[i][j] = pre[i][j] + pre[i - 1][j];
        }
      }
    }

    // Traverse the array arr[]
    for (int i = n ; i >= 2 ; --i) {

      // For each K
      for (int j = 0 ; j <= k ; j++) {

        long val = f(k, j);

        long val1 = (long)(Math.pow(A[i], k - j)) * pre[i - 1][j];

        val = val * val1;

        if (j % 2 == 0)
          ans = (ans + val);

        else
          ans = (ans - val);
      }
    }

    ans = (long)(2) * ans;

    // Return the final answer
    return ans;
  }

  // Driver code
  public static void main(String[] args){

    // Given N and K
    n = 3;
    k = 3;

    // Initializing with -1
    for(int i = 0 ; i <= 100 ; i++){
      ncr[i] = new long[101];
      for(int j = 0 ; j <= 100 ; j++){
        ncr[i][j] = -1;
      }
    }

    // Given array
    // Making vector A as 1-Indexing
    // So adding 0 in the beginning
    A = new long[]{ 0, 1, 2, 3 };

    // Function Call
    System.out.println(pairsPower());

  }
}

// This code is contributed by aadityaburujwale.

Python3

# Python3 program for the above approach
class Solution:

    def __init__(self, N, K, B):

        self.ncr = []

        # Since K can be 100 max
        for i in range(101):
            temp = []
            for j in range(101):

                # Initializing with -1
                temp.append(-1)

            self.ncr.append(temp)

        self.n = N
        self.k = K

        # Making vector A as 1-Indexing
        self.A = [0] + B

    # To Calculate the value nCk
    def f(self, n, k):

        if k == 0:
            return 1
        if n == k:
            return 1
        if n < k:
            return 0

        if self.ncr[n][k] != -1:
            return self.ncr[n][k]

        # Since nCj = (n-1)Cj + (n-1)C(j-1);
        self.ncr[n][k] = (self.f(n - 1, k) +
                          self.f(n - 1, k - 1))

        return self.ncr[n][k]

    # Function that summation of absolute
    # differences of all pairs raised
    # to the power k
    def pairsPower(self):

        pre = []

        for i in range(self.n + 1):
            temp = []
            for j in range(self.k + 1):
                temp.append(0)

            pre.append(temp)

        ans = 0

        # Sort the given array
        self.A.sort()

        # Precomputation part, O(n*k)
        for i in range(1, self.n + 1):
            pre[i][0] = 1

            for j in range(1, self.k + 1):
                pre[i][j] = (self.A[i] *
                                pre[i][j - 1])

            if i != 1:
                for j in range(self.k + 1):
                    pre[i][j] = (pre[i][j] +
                                 pre[i - 1][j])

        # Traverse the array arr[]
        for i in range(self.n, 1, -1):

            # For each K
            for j in range(self.k + 1):
                val = self.f(self.k, j)
                val1 = (pow(self.A[i],
                            self.k - j) *
                             pre[i - 1][j])

                val = val * val1

                if j % 2 == 0:
                    ans = ans + val
                else:
                    ans = ans - val

        ans = 2 * ans

        # Return the final answer
        return ans

# Driver code
if __name__ == '__main__':

    # Given N and K
    N = 3
    K = 3

    # Given array
    arr = [ 1, 2, 3 ]

    # Creation of object of class
    obj = Solution(N, K, arr)

    # Function call
    print(obj.pairsPower())

# This code is contributed by Shivam Singh

C#

// C# program to implement above approach
using System;
using System.Collections;
using System.Collections.Generic;

class GFG
{

    // Since K can be 100 max
    static long[][] ncr = new long[101][];
    static int n, k;
    static List<long> A;

    // To Calculate the value nCk
    static long f(int n, int k)
    {
        if (k == 0)
            return (long)1;

        if (n == k)
            return (long)1;

        if (n < k)
            return 0;

        if (ncr[n][k] != -1)
            return ncr[n][k];

        ncr[n][k] = f(n - 1, k) + f(n - 1, k - 1);
        // Since nCj = (n-1)Cj + (n-1)C(j-1);
        return ncr[n][k];
    }

    // Function that summation of absolute
    // differences of all pairs raised
    // to the power k
    static long pairsPower()
    {
        long[][] pre = new long[n + 1][];
        long ans = 0;
        for(int i = 0 ; i <= n ; i++){
            pre[i] = new long[k + 1];
        }

        // Sort the given array
        A.Sort(1, A.Count - 1, Comparer<long>.Default);

        // Precomputation part, O(n*k)
        for (int i = 1 ; i <= n ; ++i) {
            pre[i][0] = (long)1;

            for (int j = 1 ; j <= k ; j++) {
                pre[i][j] = A[i] * pre[i][j - 1];
            }

            if (i != 1) {
                for (int j = 0 ; j <= k ; ++j){
                    pre[i][j] = pre[i][j] + pre[i - 1][j];
                }
            }
        }

        // Traverse the array arr[]
        for (int i = n ; i >= 2 ; --i) {

            // For each K
            for (int j = 0 ; j <= k ; j++) {

                long val = f(k, j);

                long val1 = (long)(Math.Pow(A[i], k - j)) * pre[i - 1][j];

                val = val * val1;

                if (j % 2 == 0)
                    ans = (ans + val);

                else
                    ans = (ans - val);
            }
        }

        ans = (long)(2) * ans;

        // Return the final answer
        return ans;
    }

    // Driver code
    public static void Main(string[] args){

        // Given N and K
        n = 3;
        k = 3;

        // Initializing with -1
        for(int i = 0 ; i <= 100 ; i++){
            ncr[i] = new long[101];
            for(int j = 0 ; j <= 100 ; j++){
                ncr[i][j] = -1;
            }
        }

        // Given array
        // Making vector A as 1-Indexing
        // So adding 0 in the beginning
        A = new List<long>{ 0, 1, 2, 3 };

        // Function Call
        Console.WriteLine(pairsPower());

    }
}

Javascript

// Javascript program for the above approach

// To Calculate the value nCk
function f(n, k)
{
    if (k == 0)
    return 1;

    if (n == k)
    return 1;

    if (n < k)
    return 0;

    if (ncr[n][k] != -1)
    return ncr[n][k];

    ncr[n][k] = f(n - 1, k) + f(n - 1, k - 1);
    // Since nCj = (n-1)Cj + (n-1)C(j-1);
    return ncr[n][k];
}

// Function that summation of absolute
// differences of all pairs raised
// to the power k
function pairsPower()
{
    let pre=[];
    let ans=0;
    for(let i = 0 ; i <= n ; i++){
     let temp=[];
    for(let j = 0 ; j <= k; j++){
        temp.push(0);
    }
    pre.push(temp);
    }

    // Sort the given array
    A.sort();

    // Precomputation part, O(n*k)
    for (let i = 1 ; i <= n ; ++i) {
    pre[i][0] = 1;

    for (let j = 1 ; j <= k ; j++) {
        pre[i][j] = A[i] * pre[i][j - 1];
    }

    if (i != 1) {
        for (let j = 0 ; j <= k ; ++j){
        pre[i][j] = pre[i][j] + pre[i - 1][j];
        }
    }
    }

    // Traverse the array arr[]
    for (let i = n ; i >= 2 ; --i) {

    // For each K
    for (let j = 0 ; j <= k ; j++) {

        let val = f(k, j);

        let val1 = (Math.pow(A[i], k - j)) * pre[i - 1][j];

        val = val * val1;

        if (j % 2 == 0)
        ans = (ans + val);

        else
        ans = (ans - val);
    }
    }

    ans = (2) * ans;

    // Return the final answer
    return ans;
}

// Driver code


    // Given N and K
    let n = 3;
    let k = 3;

    let ncr = [];
    // Initializing with -1
    for(let i = 0 ; i <= 100 ; i++){
     let temp=[];
    for(let j = 0 ; j <= 100 ; j++){
        temp.push(-1);
    }
    ncr.push(temp);
    }

    // Given array
    // Making vector A as 1-Indexing
    // So adding 0 in the beginning
    A = [0, 1, 2, 3];

    // Function Call
   console.log(pairsPower());

// This code is contributed by Pushpesh Raj.

Output

20

Time Complexity: O(N*K)
Auxiliary Space: O(N*K)