Related Articles

# Minimize the sum after choosing elements from the given three arrays

• Difficulty Level : Medium
• Last Updated : 28 May, 2021

Given three arrays A[], B[] and C[] of same size N. The task is to minimize the sum after choosing N elements from these array such that at every index i an element from any one of the array A[i], B[i] or C[i] can be chosen and no two consecutive elements can be chosen from the same array.
Examples:

Input: A[] = {1, 100, 1}, B[] = {100, 100, 100}, C[] = {100, 100, 100}
Output: 102
A + B + A = 1 + 100 + 100 = 201
A + B + C = 1 + 100 + 100 = 201
A + C + B = 1 + 100 + 100 = 201
A + C + A = 1 + 100 + 1 = 102
B + A + B = 100 + 100 + 100 = 300
B + A + C = 100 + 100 + 100 = 300
B + C + A = 100 + 100 + 1 = 201
B + C + B = 100 + 100 + 100 = 300
C + A + B = 100 + 100 + 100 = 300
C + A + C = 100 + 100 + 100 = 300
C + B + A = 100 + 100 + 1 = 201
C + B + C = 100 + 100 + 100 = 300
Input: A[] = {1, 1, 1}, B[] = {1, 1, 1}, C[] = {1, 1, 1}
Output:

Approach: The problem is a simple variation of finding minimum cost. The extra constraint are that if we take an element from a particular array then we cannot take the next element from the same array. This could easily be solved using recursion but it would give time complexity as O(3^n) because for every element we have three arrays as choices.
To improve the time complexity we can easily store the pre-calculated values in a dp array.
Since there are three arrays to choose from at every index, three cases arise in this scenario:

• Case 1: If array A[] is selected from the ith element then we either choose the array B[] or the array C[] for the (i + 1)th element.
• Case 2: If array B[] is selected from the ith element then we either choose the array A[] or the array C[] for the (i + 1)th element.
• Case 3: If array C[] is selected from the ith element then we either choose the array A[] or the array B[] for the (i + 1)th element.

The above states can be solved using recursion and intermediate results can be stored in the dp array.
Below is the implementation of the above approach:

## C++

 `// C++ implementation of the above approach` `#include ``using` `namespace` `std;``#define SIZE 3``const` `int` `N = 3;` `// Function to return the minimized sum``int` `minSum(``int` `A[], ``int` `B[], ``int` `C[], ``int` `i,``        ``int` `n, ``int` `curr, ``int` `dp[SIZE][N])``{` `    ``// If all the indices have been used``    ``if` `(n <= 0)``        ``return` `0;` `    ``// If this value is pre-calculated``    ``// then return its value from dp array``    ``// instead of re-computing it``    ``if` `(dp[n][curr] != -1)``        ``return` `dp[n][curr];` `    ``// Here curr is the array chosen``    ``// for the (i - 1)th element``    ``// 0 for A[], 1 for B[] and 2 for C[]` `    ``// If A[i - 1] was chosen previously then``    ``// only B[i] or C[i] can chosen now``    ``// choose the one which leads``    ``// to the minimum sum``    ``if` `(curr == 0) {``        ``return` `dp[n][curr]``                ``= min(B[i] + minSum(A, B, C, i + 1, n - 1, 1, dp),``                      ``C[i] + minSum(A, B, C, i + 1, n - 1, 2, dp));``    ``}` `    ``// If B[i - 1] was chosen previously then``    ``// only A[i] or C[i] can chosen now``    ``// choose the one which leads``    ``// to the minimum sum``    ``if` `(curr == 1)``        ``return` `dp[n][curr]``                ``= min(A[i] + minSum(A, B, C, i + 1, n - 1, 0, dp),``                      ``C[i] + minSum(A, B, C, i + 1, n - 1, 2, dp));` `    ``// If C[i - 1] was chosen previously then``    ``// only A[i] or B[i] can chosen now``    ``// choose the one which leads``    ``// to the minimum sum``    ``return` `dp[n][curr]``                ``= min(A[i] + minSum(A, B, C, i + 1, n - 1, 0, dp),``                      ``B[i] + minSum(A, B, C, i + 1, n - 1, 1, dp));``}` `// Driver code``int` `main()``{``    ``int` `A[] = { 1, 50, 1 };``    ``int` `B[] = { 50, 50, 50 };``    ``int` `C[] = { 50, 50, 50 };` `    ``// Initialize the dp[][] array``    ``int` `dp[SIZE][N];``    ``for` `(``int` `i = 0; i < SIZE; i++)``        ``for` `(``int` `j = 0; j < N; j++)``            ``dp[i][j] = -1;` `    ``// min(start with A, start with B, start with C)``    ``cout << min(A + minSum(A, B, C, 1, SIZE - 1, 0, dp),``                ``min(B + minSum(A, B, C, 1, SIZE - 1, 1, dp),``                    ``C + minSum(A, B, C, 1, SIZE - 1, 2, dp)));` `    ``return` `0;``}`

## Java

 `// Java implementation of the above approach``import` `java.io.*;` `class` `GFG``{` `static` `int` `SIZE = ``3``;``static` `int` `N = ``3``;` `// Function to return the minimized sum``static` `int` `minSum(``int` `A[], ``int` `B[], ``int` `C[], ``int` `i,``                    ``int` `n, ``int` `curr, ``int` `[][]dp)``{` `    ``// If all the indices have been used``    ``if` `(n <= ``0``)``        ``return` `0``;` `    ``// If this value is pre-calculated``    ``// then return its value from dp array``    ``// instead of re-computing it``    ``if` `(dp[n][curr] != -``1``)``        ``return` `dp[n][curr];` `    ``// Here curr is the array chosen``    ``// for the (i - 1)th element``    ``// 0 for A[], 1 for B[] and 2 for C[]` `    ``// If A[i - 1] was chosen previously then``    ``// only B[i] or C[i] can chosen now``    ``// choose the one which leads``    ``// to the minimum sum``    ``if` `(curr == ``0``)``    ``{``        ``return` `dp[n][curr]``                ``= Math.min(B[i] + minSum(A, B, C, i + ``1``, n - ``1``, ``1``, dp),``                    ``C[i] + minSum(A, B, C, i + ``1``, n - ``1``, ``2``, dp));``    ``}` `    ``// If B[i - 1] was chosen previously then``    ``// only A[i] or C[i] can chosen now``    ``// choose the one which leads``    ``// to the minimum sum``    ``if` `(curr == ``1``)``        ``return` `dp[n][curr]``                ``= Math.min(A[i] + minSum(A, B, C, i + ``1``, n - ``1``, ``0``, dp),``                    ``C[i] + minSum(A, B, C, i + ``1``, n - ``1``, ``2``, dp));` `    ``// If C[i - 1] was chosen previously then``    ``// only A[i] or B[i] can chosen now``    ``// choose the one which leads``    ``// to the minimum sum``    ``return` `dp[n][curr]``                ``= Math.min(A[i] + minSum(A, B, C, i + ``1``, n - ``1``, ``0``, dp),``                    ``B[i] + minSum(A, B, C, i + ``1``, n - ``1``, ``1``, dp));``}` `// Driver code``public` `static` `void` `main (String[] args)``{``    ``int` `A[] = { ``1``, ``50``, ``1` `};``    ``int` `B[] = { ``50``, ``50``, ``50` `};``    ``int` `C[] = { ``50``, ``50``, ``50` `};``    ` `    ``// Initialize the dp[][] array``    ``int` `dp[][] = ``new` `int``[SIZE][N];``    ``for` `(``int` `i = ``0``; i < SIZE; i++)``        ``for` `(``int` `j = ``0``; j < N; j++)``            ``dp[i][j] = -``1``;``    ` `    ``// min(start with A, start with B, start with C)``    ``System.out.println(Math.min(A[``0``] + minSum(A, B, C, ``1``, SIZE - ``1``, ``0``, dp),``                ``Math.min(B[``0``] + minSum(A, B, C, ``1``, SIZE - ``1``, ``1``, dp),``                    ``C[``0``] + minSum(A, B, C, ``1``, SIZE - ``1``, ``2``, dp))));``}``}` `// This code is contributed by anuj_67..`

## Python3

 `# Python3 implementation of the above approach` `import` `numpy as np` `SIZE ``=` `3``;``N ``=` `3``;` `# Function to return the minimized sum``def` `minSum(A, B, C, i, n, curr, dp) :` `    ``# If all the indices have been used``    ``if` `(n <``=` `0``) :``        ``return` `0``;` `    ``# If this value is pre-calculated``    ``# then return its value from dp array``    ``# instead of re-computing it``    ``if` `(dp[n][curr] !``=` `-``1``) :``        ``return` `dp[n][curr];` `    ``# Here curr is the array chosen``    ``# for the (i - 1)th element``    ``# 0 for A[], 1 for B[] and 2 for C[]` `    ``# If A[i - 1] was chosen previously then``    ``# only B[i] or C[i] can chosen now``    ``# choose the one which leads``    ``# to the minimum sum``    ``if` `(curr ``=``=` `0``) :``        ``dp[n][curr] ``=` `min``( B[i] ``+` `minSum(A, B, C, i ``+` `1``, n ``-` `1``, ``1``, dp),``        ``C[i] ``+` `minSum(A, B, C, i ``+` `1``, n ``-` `1``, ``2``, dp));``        ``return` `dp[n][curr]``    `  `    ``# If B[i - 1] was chosen previously then``    ``# only A[i] or C[i] can chosen now``    ``# choose the one which leads``    ``# to the minimum sum``    ``if` `(curr ``=``=` `1``) :``        ``dp[n][curr] ``=` `min``(A[i] ``+` `minSum(A, B, C, i ``+` `1``, n ``-` `1``, ``0``, dp),``        ``C[i] ``+` `minSum(A, B, C, i ``+` `1``, n ``-` `1``, ``2``, dp));``        ``return` `dp[n][curr]` `    ``# If C[i - 1] was chosen previously then``    ``# only A[i] or B[i] can chosen now``    ``# choose the one which leads``    ``# to the minimum sum``    ``dp[n][curr] ``=` `min``(A[i] ``+` `minSum(A, B, C, i ``+` `1``, n ``-` `1``, ``0``, dp),``    ``B[i] ``+` `minSum(A, B, C, i ``+` `1``, n ``-` `1``, ``1``, dp));``    ` `    ``return` `dp[n][curr]`  `# Driver code``if` `__name__ ``=``=` `"__main__"` `:` `    ``A ``=` `[ ``1``, ``50``, ``1` `];``    ``B ``=` `[ ``50``, ``50``, ``50` `];``    ``C ``=` `[ ``50``, ``50``, ``50` `];` `    ``# Initialize the dp[][] array``    ``dp ``=` `np.zeros((SIZE,N));``    ` `    ``for` `i ``in` `range``(SIZE) :``        ``for` `j ``in` `range``(N) :``            ``dp[i][j] ``=` `-``1``;` `    ``# min(start with A, start with B, start with C)``    ``print``(``min``(A[``0``] ``+` `minSum(A, B, C, ``1``, SIZE ``-` `1``, ``0``, dp),``                ``min``(B[``0``] ``+` `minSum(A, B, C, ``1``, SIZE ``-` `1``, ``1``, dp),``                ``C[``0``] ``+` `minSum(A, B, C, ``1``, SIZE ``-` `1``, ``2``, dp))));` `# This code is contributed by AnkitRai01`

## C#

 `// C# implementation of the above approach``using` `System;` `class` `GFG``{` `static` `int` `SIZE = 3;``static` `int` `N = 3;` `// Function to return the minimized sum``static` `int` `minSum(``int` `[]A, ``int` `[]B, ``int` `[]C, ``int` `i,``                    ``int` `n, ``int` `curr, ``int` `[,]dp)``{` `    ``// If all the indices have been used``    ``if` `(n <= 0)``        ``return` `0;` `    ``// If this value is pre-calculated``    ``// then return its value from dp array``    ``// instead of re-computing it``    ``if` `(dp[n,curr] != -1)``        ``return` `dp[n,curr];` `    ``// Here curr is the array chosen``    ``// for the (i - 1)th element``    ``// 0 for A[], 1 for B[] and 2 for C[]` `    ``// If A[i - 1] was chosen previously then``    ``// only B[i] or C[i] can chosen now``    ``// choose the one which leads``    ``// to the minimum sum``    ``if` `(curr == 0)``    ``{``        ``return` `dp[n,curr]``                ``= Math.Min(B[i] + minSum(A, B, C, i + 1, n - 1, 1, dp),``                    ``C[i] + minSum(A, B, C, i + 1, n - 1, 2, dp));``    ``}` `    ``// If B[i - 1] was chosen previously then``    ``// only A[i] or C[i] can chosen now``    ``// choose the one which leads``    ``// to the minimum sum``    ``if` `(curr == 1)``        ``return` `dp[n,curr]``                ``= Math.Min(A[i] + minSum(A, B, C, i + 1, n - 1, 0, dp),``                    ``C[i] + minSum(A, B, C, i + 1, n - 1, 2, dp));` `    ``// If C[i - 1] was chosen previously then``    ``// only A[i] or B[i] can chosen now``    ``// choose the one which leads``    ``// to the minimum sum``    ``return` `dp[n,curr]``                ``= Math.Min(A[i] + minSum(A, B, C, i + 1, n - 1, 0, dp),``                    ``B[i] + minSum(A, B, C, i + 1, n - 1, 1, dp));``}` `// Driver code``public` `static` `void` `Main ()``{``    ``int` `[]A = { 1, 50, 1 };``    ``int` `[]B = { 50, 50, 50 };``    ``int` `[]C = { 50, 50, 50 };``    ` `    ``// Initialize the dp[][] array``    ``int` `[,]dp = ``new` `int``[SIZE,N];``    ``for` `(``int` `i = 0; i < SIZE; i++)``        ``for` `(``int` `j = 0; j < N; j++)``            ``dp[i,j] = -1;``    ` `    ``// min(start with A, start with B, start with C)``    ``Console.WriteLine(Math.Min(A + minSum(A, B, C, 1, SIZE - 1, 0, dp),``                ``Math.Min(B + minSum(A, B, C, 1, SIZE - 1, 1, dp),``                    ``C + minSum(A, B, C, 1, SIZE - 1, 2, dp))));``}``}` `// This code is contributed by anuj_67..`

## Javascript

 ``
Output:
`52`

Attention reader! Don’t stop learning now. Get hold of all the important DSA concepts with the DSA Self Paced Course at a student-friendly price and become industry ready.  To complete your preparation from learning a language to DS Algo and many more,  please refer Complete Interview Preparation Course.

In case you wish to attend live classes with experts, please refer DSA Live Classes for Working Professionals and Competitive Programming Live for Students.

My Personal Notes arrow_drop_up