Horje
Minimum Distance between Runners before the Race Ends

Given two integers N, K and two arrays speed[] and dist[] each of size N, where speed[i] is the speed of the ith runner and dist[i] is the distance which the ith runner is ahead of the starting line, find the minimum value between the most distant runners at any time if the race ends after K seconds. In other words, we have to tell the minimum distance between the two most distant runners at any time until the race ends.

Examples:

Input: N = 3, speed[] = {1, 2, 3}, dist[] = {1, 2, 3}, K = 3
Output: 2
Explanation: The minimum distance between the two most distant runners will be 2 at time t = 0. As soon as time increases, the distance between the first and the last runner(most distant runners) will keep on increasing.

Input: N = 3, speed[] = {1, 2, 3}, dist[] = {3, 2, 1}, K = 3
Output: 0
Explanation: The minimum distance between the two most distant runners will be 0 at time t = 1.

  • At time t = 0, dist[] = {3, 2, 1}, distance between most distant runners = 3 – 1 = 2
  • At time t = 1, dist[] = {4, 4, 4}, distance between most distant runners = 4 – 4 = 0
  • At time t = 2, dist[] = {5, 6, 7}, distance between most distant runners = 7 – 5 = 2
  • At time t = 3, dist[] = {6, 8, 10}, distance between most distant runners = 10 – 6 = 4

Approach: To solve the problem, follow the below idea:

The problem can be solved using Ternary Search as the minimum distance between two most distant runners firstly decreases and then increases. The minimum distance between the most distant runners behaves as a Unimodal function, therefore we can apply Ternary Search. So, we can calculate two mid points and calculate the distance between the most distant runners at those two mid points. Now, according to the minimum distance we can reduce the search space. We can initialize the low as 0 and high as K to find the time at which the distance between the most distant runners is minimum. Let f(x) be the minimum distance between two most distant runners at time = x. So,

  • If f(mid1) < f(mid2), then hi = mid2.
  • If f(mid1) == f(mid2), then lo = mid1 and hi = mid2.
  • If f(mid1) > f(mid2), then lo = mid1.

Now, we can keep reducing the search space till hi – lo > 2 and then calculate the answer for time between lo and hi.

Step-by-step algorithm:

  • Initialize the search space using lo = 0, hi = K.
  • Calculate two mid points, mid1 and mid2 and find the distance between the most distant runners at time = mid1, say dist1 and time = mid2, say dist2.
  • If dist1 == dist2
    • lo = mid1 and hi = mid2
  • Else If dist1 < dist2
    • hi = mid2
  • Else if dist1 > dist2
    • lo = mid1
  • Reduce the search space till 300 times to get better precision.
  • Return the final answer as the minimum distance between two runners at time = hi or lo.

Below is the implementation of the approach:

C++

#include <bits/stdc++.h>
using namespace std;
 
// function to calculate the distance between the most
// distant runners at time = mid
float f(float mid, int N, int* speed, int* dist)
{
    float minDist = 1e9, maxDist = -1e9;
    for (int i = 0; i < N; i++) {
        float currDist = dist[i] + mid * (speed[i]);
        minDist = min(minDist, currDist);
        maxDist = max(maxDist, currDist);
    }
    return maxDist - minDist;
}
 
// function to calculate the minimum distance between the
// most distant runners from t = 0 to t = K
float getMinimumDistance(int N, int K, int* speed,
                        int* dist)
{
    float lo = 0, hi = K;
    // 300 operations are enough for precision
    int cnt = 300;
    while (cnt != 0) {
        cnt--;
        float mid1 = lo + (hi - lo) / 3.0;
        float mid2 = hi - (hi - lo) / 3.0;
 
        float dist1 = f(mid1, N, speed, dist);
        float dist2 = f(mid2, N, speed, dist);
        if (dist1 == dist2) {
            lo = mid1;
            hi = mid2;
        }
        else if (dist1 < dist2) {
            hi = mid2;
        }
        else {
            lo = mid1;
        }
    }
    return f(hi, N, speed, dist);
}
 
int main()
{
 
    int N = 3, K = 3;
    int speed[] = { 1, 2, 3 }, dist[] = { 1, 2, 3 };
 
    cout << setprecision(6)
        << getMinimumDistance(N, K, speed, dist) << "\n";
    return 0;
}

Java

import java.util.Arrays;
 
public class Solution {
    // Function to calculate the distance between the most
    // distant runners at time = mid
    static float f(float mid, int N, int[] speed, int[] dist) {
        float minDist = 1e9f, maxDist = -1e9f;
        for (int i = 0; i < N; i++) {
            float currDist = dist[i] + mid * speed[i];
            minDist = Math.min(minDist, currDist);
            maxDist = Math.max(maxDist, currDist);
        }
        return maxDist - minDist;
    }
 
    // Function to calculate the minimum distance between the
    // most distant runners from t = 0 to t = K
    static float getMinimumDistance(int N, int K, int[] speed, int[] dist) {
        float lo = 0, hi = K;
        // 300 operations are enough for precision
        int cnt = 300;
        while (cnt != 0) {
            cnt--;
            float mid1 = lo + (hi - lo) / 3.0f;
            float mid2 = hi - (hi - lo) / 3.0f;
 
            float dist1 = f(mid1, N, speed, dist);
            float dist2 = f(mid2, N, speed, dist);
            if (dist1 == dist2) {
                lo = mid1;
                hi = mid2;
            } else if (dist1 < dist2) {
                hi = mid2;
            } else {
                lo = mid1;
            }
        }
        return f(hi, N, speed, dist);
    }
 
    public static void main(String[] args) {
        int N = 3, K = 3;
        int[] speed = {1, 2, 3};
        int[] dist = {1, 2, 3};
 
        System.out.println(String.format("%.6f", getMinimumDistance(N, K, speed, dist)));
    }
   
}
 
// This code is contributed by shivamgupta0987654321

Python3

# function to calculate the distance between the most
# distant runners at time = mid
def f(mid, N, speed, dist):
    minDist = 1e9
    maxDist = -1e9
    for i in range(N):
        currDist = dist[i] + mid * speed[i]
        minDist = min(minDist, currDist)
        maxDist = max(maxDist, currDist)
    return maxDist - minDist
 
# function to calculate the minimum distance between the
# most distant runners from t = 0 to t = K
def get_minimum_distance(N, K, speed, dist):
    lo = 0
    hi = K
    # 300 operations are enough for precision
    cnt = 300
    while cnt != 0:
        cnt -= 1
        mid1 = lo + (hi - lo) / 3.0
        mid2 = hi - (hi - lo) / 3.0
 
        dist1 = f(mid1, N, speed, dist)
        dist2 = f(mid2, N, speed, dist)
        if dist1 == dist2:
            lo = mid1
            hi = mid2
        elif dist1 < dist2:
            hi = mid2
        else:
            lo = mid1
 
    return f(hi, N, speed, dist)
 
if __name__ == "__main__":
    N = 3
    K = 3
    speed = [1, 2, 3]
    dist = [1, 2, 3]
 
    print("{:.6f}".format(get_minimum_distance(N, K, speed, dist)))
 
# This code is contributed by shivamgupta0987654321

C#

using System;
 
public class Solution
{
    // Function to calculate the distance between the most
    // distant runners at time = mid
    static float f(float mid, int N, int[] speed, int[] dist)
    {
        float minDist = 1e9f, maxDist = -1e9f;
        for (int i = 0; i < N; i++)
        {
            float currDist = dist[i] + mid * speed[i];
            minDist = Math.Min(minDist, currDist);
            maxDist = Math.Max(maxDist, currDist);
        }
        return maxDist - minDist;
    }
 
    // Function to calculate the minimum distance between the
    // most distant runners from t = 0 to t = K
    static float GetMinimumDistance(int N, int K, int[] speed, int[] dist)
    {
        float lo = 0, hi = K;
        // 300 operations are enough for precision
        int cnt = 300;
        while (cnt != 0)
        {
            cnt--;
            float mid1 = lo + (hi - lo) / 3.0f;
            float mid2 = hi - (hi - lo) / 3.0f;
 
            float dist1 = f(mid1, N, speed, dist);
            float dist2 = f(mid2, N, speed, dist);
            if (dist1 == dist2)
            {
                lo = mid1;
                hi = mid2;
            }
            else if (dist1 < dist2)
            {
                hi = mid2;
            }
            else
            {
                lo = mid1;
            }
        }
        return f(hi, N, speed, dist);
    }
 
    public static void Main(string[] args)
    {
        int N = 3, K = 3;
        int[] speed = { 1, 2, 3 };
        int[] dist = { 1, 2, 3 };
 
        Console.WriteLine(string.Format("{0:F6}", GetMinimumDistance(N, K, speed, dist)));
    }
}

Javascript

// function to calculate the distance between the most
// distant runners at time = mid
function f(mid, N, speed, dist) {
    let minDist = 1e9, maxDist = -1e9;
    for (let i = 0; i < N; i++) {
        let currDist = dist[i] + mid * speed[i];
        minDist = Math.min(minDist, currDist);
        maxDist = Math.max(maxDist, currDist);
    }
    return maxDist - minDist;
}
 
// function to calculate the minimum distance between the
// most distant runners from t = 0 to t = K
function getMinimumDistance(N, K, speed, dist) {
    let lo = 0, hi = K;
    // 300 operations are enough for precision
    let cnt = 300;
    while (cnt !== 0) {
        cnt--;
        let mid1 = lo + (hi - lo) / 3.0;
        let mid2 = hi - (hi - lo) / 3.0;
 
        let dist1 = f(mid1, N, speed, dist);
        let dist2 = f(mid2, N, speed, dist);
        if (dist1 === dist2) {
            lo = mid1;
            hi = mid2;
        }
        else if (dist1 < dist2) {
            hi = mid2;
        }
        else {
            lo = mid1;
        }
    }
    return f(hi, N, speed, dist);
}
 
// main function
function main() {
    let N = 3, K = 3;
    let speed = [1, 2, 3], dist = [1, 2, 3];
 
    console.log(getMinimumDistance(N, K, speed, dist).toFixed(6));
}
 
main();

Output

2

Time Complexity: O(N * logK), where N is the number of runners and K is the duration of the race.
Auxiliary Space: O(1)


 


Reffered: https://www.geeksforgeeks.org

Competitive Programming

Related
Why do you do Competitive Programming (CP)? Why do you do Competitive Programming (CP)?
10 reasons not to quit Competitive Programming 10 reasons not to quit Competitive Programming
Number Theory for DSA &amp; Competitive Programming Number Theory for DSA &amp; Competitive Programming
Segment Tree for Range Assignment and Range Sum Queries Segment Tree for Range Assignment and Range Sum Queries
Segment Tree for Range Multiplication and Sum Queries Segment Tree for Range Multiplication and Sum Queries

Type:
 Geek
Category:
 Coding
Sub Category:
 Tutorial
Uploaded by:
 Admin
Views:
 13



 

Horje