1425. Constrained Subsequence Sum

Description

Given an integer array nums and an integer k, return the maximum sum of a non-empty subsequence of that array such that for every two consecutive integers in the subsequence, nums[i] and nums[j], where i < j, the condition j - i <= k is satisfied.

A subsequence of an array is obtained by deleting some number of elements (can be zero) from the array, leaving the remaining elements in their original order.

Example 1:

Input: nums = [10,2,-10,5,20], k = 2
Output: 37
Explanation: The subsequence is [10, 2, 5, 20].

Example 2:

Input: nums = [-1,-2,-3], k = 1
Output: -1
Explanation: The subsequence must be non-empty, so we choose the largest number.

Example 3:

Input: nums = [10,-2,-10,-5,20], k = 2
Output: 23
Explanation: The subsequence is [10, -2, -5, 20].

Constraints:

1 <= k <= nums.length <= 10⁵
-10⁴ <= nums[i] <= 10⁴

Solutions

Solution 1: Dynamic Programming + Monotonic Queue

We define \(f[i]\) to represent the maximum sum of the subsequence ending at \(\textit{nums}[i]\) that meets the conditions. Initially, \(f[i] = 0\), and the answer is \(\max_{0 \leq i \lt n} f(i)\).

We notice that the problem requires us to maintain the maximum value of a sliding window, which is a typical application scenario for a monotonic queue. We can use a monotonic queue to optimize the dynamic programming transition.

We maintain a monotonic queue \(q\) that is decreasing from the front to the back, storing the indices \(i\). Initially, we add a sentinel \(0\) to the queue.

We traverse \(i\) from \(0\) to \(n - 1\). For each \(i\), we perform the following operations:

If the front element \(q[0]\) satisfies \(i - q[0] > k\), it means the front element is no longer within the sliding window, and we need to remove the front element from the queue;
Then, we calculate \(f[i] = \max(0, f[q[0]]) + \textit{nums}[i]\), which means we add \(\textit{nums}[i]\) to the sliding window to get the maximum subsequence sum;
Next, we update the answer \(\textit{ans} = \max(\textit{ans}, f[i])\);
Finally, we add \(i\) to the back of the queue and maintain the monotonicity of the queue. If \(f[q[\textit{back}]] \leq f[i]\), we need to remove the back element until the queue is empty or \(f[q[\textit{back}]] > f[i]\).

The final answer is \(\textit{ans}\).

The time complexity is \(O(n)\), and the space complexity is \(O(n)\). Here, \(n\) is the length of the array \(\textit{nums}\).

Python3JavaC++GoTypeScript

class Solution:
    def constrainedSubsetSum(self, nums: List[int], k: int) -> int:
        q = deque([0])
        n = len(nums)
        f = [0] * n
        ans = -inf
        for i, x in enumerate(nums):
            while i - q[0] > k:
                q.popleft()
            f[i] = max(0, f[q[0]]) + x
            ans = max(ans, f[i])
            while q and f[q[-1]] <= f[i]:
                q.pop()
            q.append(i)
        return ans

class Solution {
    public int constrainedSubsetSum(int[] nums, int k) {
        Deque<Integer> q = new ArrayDeque<>();
        q.offer(0);
        int n = nums.length;
        int[] f = new int[n];
        int ans = -(1 << 30);
        for (int i = 0; i < n; ++i) {
            while (i - q.peekFirst() > k) {
                q.pollFirst();
            }
            f[i] = Math.max(0, f[q.peekFirst()]) + nums[i];
            ans = Math.max(ans, f[i]);
            while (!q.isEmpty() && f[q.peekLast()] <= f[i]) {
                q.pollLast();
            }
            q.offerLast(i);
        }
        return ans;
    }
}

class Solution {
public:
    int constrainedSubsetSum(vector<int>& nums, int k) {
        deque<int> q = {0};
        int n = nums.size();
        int f[n];
        f[0] = 0;
        int ans = INT_MIN;
        for (int i = 0; i < n; ++i) {
            while (i - q.front() > k) {
                q.pop_front();
            }
            f[i] = max(0, f[q.front()]) + nums[i];
            ans = max(ans, f[i]);
            while (!q.empty() && f[q.back()] <= f[i]) {
                q.pop_back();
            }
            q.push_back(i);
        }
        return ans;
    }
};

func constrainedSubsetSum(nums []int, k int) int {
    q := Deque{}
    q.PushFront(0)
    n := len(nums)
    f := make([]int, n)
    ans := nums[0]
    for i, x := range nums {
        for i-q.Front() > k {
            q.PopFront()
        }
        f[i] = max(0, f[q.Front()]) + x
        ans = max(ans, f[i])
        for !q.Empty() && f[q.Back()] <= f[i] {
            q.PopBack()
        }
        q.PushBack(i)
    }
    return ans
}

// template
type Deque struct{ l, r []int }

func (q Deque) Empty() bool {
    return len(q.l) == 0 && len(q.r) == 0
}

func (q Deque) Size() int {
    return len(q.l) + len(q.r)
}

func (q *Deque) PushFront(v int) {
    q.l = append(q.l, v)
}

func (q *Deque) PushBack(v int) {
    q.r = append(q.r, v)
}

func (q *Deque) PopFront() (v int) {
    if len(q.l) > 0 {
        q.l, v = q.l[:len(q.l)-1], q.l[len(q.l)-1]
    } else {
        v, q.r = q.r[0], q.r[1:]
    }
    return
}

func (q *Deque) PopBack() (v int) {
    if len(q.r) > 0 {
        q.r, v = q.r[:len(q.r)-1], q.r[len(q.r)-1]
    } else {
        v, q.l = q.l[0], q.l[1:]
    }
    return
}

func (q Deque) Front() int {
    if len(q.l) > 0 {
        return q.l[len(q.l)-1]
    }
    return q.r[0]
}

func (q Deque) Back() int {
    if len(q.r) > 0 {
        return q.r[len(q.r)-1]
    }
    return q.l[0]
}

func (q Deque) Get(i int) int {
    if i < len(q.l) {
        return q.l[len(q.l)-1-i]
    }
    return q.r[i-len(q.l)]
}

function constrainedSubsetSum(nums: number[], k: number): number {
    const q = new Deque<number>();
    const n = nums.length;
    q.pushBack(0);
    let ans = nums[0];
    const f: number[] = Array(n).fill(0);
    for (let i = 0; i < n; ++i) {
        while (i - q.frontValue()! > k) {
            q.popFront();
        }
        f[i] = Math.max(0, f[q.frontValue()!]!) + nums[i];
        ans = Math.max(ans, f[i]);
        while (!q.isEmpty() && f[q.backValue()!]! <= f[i]) {
            q.popBack();
        }
        q.pushBack(i);
    }
    return ans;
}

class Node<T> {
    value: T;
    next: Node<T> | null;
    prev: Node<T> | null;

    constructor(value: T) {
        this.value = value;
        this.next = null;
        this.prev = null;
    }
}

class Deque<T> {
    private front: Node<T> | null;
    private back: Node<T> | null;
    private size: number;

    constructor() {
        this.front = null;
        this.back = null;
        this.size = 0;
    }

    pushFront(val: T): void {
        const newNode = new Node(val);
        if (this.isEmpty()) {
            this.front = newNode;
            this.back = newNode;
        } else {
            newNode.next = this.front;
            this.front!.prev = newNode;
            this.front = newNode;
        }
        this.size++;
    }

    pushBack(val: T): void {
        const newNode = new Node(val);
        if (this.isEmpty()) {
            this.front = newNode;
            this.back = newNode;
        } else {
            newNode.prev = this.back;
            this.back!.next = newNode;
            this.back = newNode;
        }
        this.size++;
    }

    popFront(): T | undefined {
        if (this.isEmpty()) {
            return undefined;
        }
        const value = this.front!.value;
        this.front = this.front!.next;
        if (this.front !== null) {
            this.front.prev = null;
        } else {
            this.back = null;
        }
        this.size--;
        return value;
    }

    popBack(): T | undefined {
        if (this.isEmpty()) {
            return undefined;
        }
        const value = this.back!.value;
        this.back = this.back!.prev;
        if (this.back !== null) {
            this.back.next = null;
        } else {
            this.front = null;
        }
        this.size--;
        return value;
    }

    frontValue(): T | undefined {
        return this.front?.value;
    }

    backValue(): T | undefined {
        return this.back?.value;
    }

    getSize(): number {
        return this.size;
    }

    isEmpty(): boolean {
        return this.size === 0;
    }
}

1425. Constrained Subsequence Sum

Description

Solutions

Solution 1: Dynamic Programming + Monotonic Queue

Comments