You are given an integer array nums and an integer k. You are asked to distribute this array into k subsets of equal size such that there are no two equal elements in the same subset.
A subset's incompatibility is the difference between the maximum and minimum elements in that array.
Return the minimum possible sum of incompatibilities of the ksubsets after distributing the array optimally, or return -1 if it is not possible.
A subset is a group integers that appear in the array with no particular order.
Example 1:
Input: nums = [1,2,1,4], k = 2
Output: 4
Explanation: The optimal distribution of subsets is [1,2] and [1,4].
The incompatibility is (2-1) + (4-1) = 4.
Note that [1,1] and [2,4] would result in a smaller sum, but the first subset contains 2 equal elements.
Example 2:
Input: nums = [6,3,8,1,3,1,2,2], k = 4
Output: 6
Explanation: The optimal distribution of subsets is [1,2], [2,3], [6,8], and [1,3].
The incompatibility is (2-1) + (3-2) + (8-6) + (3-1) = 6.
Example 3:
Input: nums = [5,3,3,6,3,3], k = 3
Output: -1
Explanation: It is impossible to distribute nums into 3 subsets where no two elements are equal in the same subset.
Constraints:
1 <= k <= nums.length <= 16
nums.length is divisible by k
1 <= nums[i] <= nums.length
Solutions
Solution 1: Preprocessing + State Compression + Dynamic Programming
Thinking
Partition into \(k\) groups of distinct values, minimizing the sum of \((\max-\min)\). \(n \le 16\) allows subset DP.
Precompute incompatibility \(g[i]\) for every size-\(m=n/k\) subset without duplicates. \(f[S]\) is the min cost of chosen index set \(S\); transition over duplicate-free \(m\)-subsets of the complement.
If \(f[2^n-1]\) stays infinite, return \(-1\).
Let's assume that the size of each subset after partitioning is \(m\), so \(m=\frac{n}{k}\), where \(n\) is the length of the array.
We can enumerate all subsets \(i\), where \(i \in [0, 2^n)\), if the binary representation of subset \(i\) has \(m\) ones, and the elements in subset \(i\) are not repeated, then we can calculate the incompatibility of subset \(i\), denoted as \(g[i]\), i.e., \(g[i]=\max_{j \in i} \{\textit{nums}[j]\} - \min_{j \in i} \{\textit{nums}[j]\}\).
Next, we can use dynamic programming to solve.
We define \(f[i]\) as the minimum sum of incompatibilities when the current partitioned subset state is \(i\). Initially, \(f[0]=0\), which means no elements are partitioned into the subset, and the rest \(f[i]=+\infty\).
For state \(i\), we find all undivided and non-repeated elements, represented by a state \(\textit{mask}\). If the number of elements in state \(\textit{mask}\) is greater than or equal to \(m\), then we enumerate all subsets \(j\) of \(\textit{mask}\), and satisfy \(j \subset \textit{mask}\), then \(f[i \cup j]=\min \{f[i \cup j], f[i]+g[j]\}\).
Finally, if \(f[2^n-1]=+\infty\), it means that it cannot be partitioned into \(k\) subsets, return \(-1\), otherwise return \(f[2^n-1]\).
The time complexity is \(O(3^n)\), and the space complexity is \(O(2^n)\), where \(n\) is the length of the array.