Блог пользователя vovuh

Автор vovuh, история, 5 лет назад, перевод, По-русски

1133A - Середина контеста

Идея: MikeMirzayanov

Разбор
Решение

1133B - Подготовка к 8 марта

Идея: MikeMirzayanov

Разбор
Решение

1133C - Сбалансированная команда

Идея: MikeMirzayanov

Разбор
Решение

1133D - Максимизация количества нулей

Идея: BledDest

Разбор
Решение

1133E - K сбалансированных команд

Идея: MikeMirzayanov

Разбор
Решение

1133F1 - Покрывающее дерево максимальной степени

Идея: MikeMirzayanov

Разбор
Решение

1133F2 - Покрывающее дерево с одной фиксированной степенью

Идея: MikeMirzayanov

Разбор
Решение
Разбор задач Codeforces Round 544 (Div. 3)
  • Проголосовать: нравится
  • +33
  • Проголосовать: не нравится

»
5 лет назад, # |
  Проголосовать: нравится +32 Проголосовать: не нравится

Explanation of f2 is copy-paste of f1, please fix it

»
5 лет назад, # |
  Проголосовать: нравится +10 Проголосовать: не нравится

F2 tutorial is missing (F1 has been repeated). Also, The problem F1 really should've been ahead of the problem E (if you want to sort the problems according to their difficulties).

»
5 лет назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

vovuh , F2 tutorial is missing

»
5 лет назад, # |
  Проголосовать: нравится +1 Проголосовать: не нравится

Can F2 be solved by finding articulation edges / bridges in the graph and as soon as we find bridge edge we merge them and after that we pick any random adjacent edges of vertex 1 so as to make its degree exactly d. and after make the whole tree connected using kruskal type algorithm? Here is my submission for reference of what i am talking. I am getting wrong answer on test 37. Pls tell if i am wrong. Thanks

  • »
    »
    5 лет назад, # ^ |
      Проголосовать: нравится +14 Проголосовать: не нравится

    The idea for F2 is that if we remove vertex 1 then the graph will be K components and if K > D there is NO answer otherwise we should connect this K components with D edges after that we can choose any MST of the remaining graph.

  • »
    »
    5 лет назад, # ^ |
      Проголосовать: нравится +3 Проголосовать: не нравится

    hope this test case will help you to understand why you are getting wa :) 7 8 4 1 2 1 3 1 4 1 5 1 6 1 7 7 6 5 4

  • »
    »
    5 лет назад, # ^ |
      Проголосовать: нравится 0 Проголосовать: не нравится

    The problem is that bridges tell us only that if that edge is removed,the graph will become disconnected, if u remove some set of edges which are not bridges it is still possible that the graph may become disconnected, that is why the above logic doesn't work.

»
5 лет назад, # |
  Проголосовать: нравится +4 Проголосовать: не нравится

Turns out that my solution for D with maps having long double keys passed. I checked the test cases and would suggest the problem setters to do the same too. From what it appears, the cases seem weird. I've linked a screenshot regarding my concerns.

»
5 лет назад, # |
Rev. 2   Проголосовать: нравится 0 Проголосовать: не нравится

why does this solution gets TLE whereas this solution gets Accepted?

»
5 лет назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

MikeMirzayanov you Have Mistakenly provide E question tutorial in F question. Can you please explain 1133F2 — Spanning Tree with One Fixed Degree solution idea.

»
5 лет назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

"How to do transitions from dpi,j? The first transition is pretty intuitive: just skip the i-th (0-indexed) student. Then we can set dp[i+1][j+1]:=max(dp[i+1][j+1],dp[i,j])."

i think it should be dp[i+1][j] as by skipping ith student we are not increasing size. MikeMirzayanov

  • »
    »
    5 лет назад, # ^ |
    Rev. 2   Проголосовать: нравится 0 Проголосовать: не нравится

    I think it's a typo. It should be dp[i+1][j+1] = max(dp[i+1][j+1],dp[i][j+1])

  • »
    »
    5 лет назад, # ^ |
      Проголосовать: нравится 0 Проголосовать: не нравится

    I also stumbled upon this, will wait for the clarification of right transition..

    • »
      »
      »
      5 лет назад, # ^ |
      Rev. 3   Проголосовать: нравится +1 Проголосовать: не нравится

      I'll will put the way I interpreted it.

      When you are in a given state (i , j) (where the students in [1,...,i] are the possible students you can use and 'j' is the maximum quantity of teams you can use) you have to take the best between pick or not the student 'i'.

      When you dont't pick student 'i' you will have to pick the solution for the case in which you have the students in [1,...,(i — 1)] as possible students to take and 'j' as the maximum quantity of teams you can use. This describe exactly what is in the state (i — 1 , j).

      When you pick student 'i' you have to see what is the best way to pick it. So you look at the states when you allow (j — 1) teams as maximum quantity of teams and students in [1,...,z] (with z <= i) as possible students (states of the form (z , j — 1) with z <= i). When you pick a state like that you could create a new team and put the maximum quantity of possible students (which are in [(z + 1),...,i]) in it to get a candidate for the state (i , j). But remember that you want the student 'i' to be in it. So you want to see only states (z , j — 1) (with z <= i) such that you could create a new team and put all students in [(z + 1),...,i] in it.

      With a good care this can be done in O(n*k): 51128813

»
5 лет назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

I will fix F2 tutorial in a few hours. Please wait a bit.

»
5 лет назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

In D, why we should ensure numerator is non-negative?

»
5 лет назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

@Vovuh How is ur editorial of F2 different from this? . Aren't they same?

»
5 лет назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

In div2 D, Why does use of unordered_map gives a TLE but normal map works fine ? Solution 51071398 gives TLE with unordered_map but this 51071641 gets Accepted with map

  • »
    »
    5 лет назад, # ^ |
      Проголосовать: нравится 0 Проголосовать: не нравится

    unordered_map has O(1) complexity on average, but input can be engineered so that it take O(n) collisions for each access. Read more about it here.

»
5 лет назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

How to do problem C using binary search?

  • »
    »
    5 лет назад, # ^ |
      Проголосовать: нравится 0 Проголосовать: не нравится

    Sort the array in ascending order. For every element in the array (let it be arr) find how many elements satisfy  ≥ arri and  ≤ arri + 5. Take maximum length of all such ranges. See this code for the implementation.

»
5 лет назад, # |
Rev. 2   Проголосовать: нравится 0 Проголосовать: не нравится

For problem D we can prevent precision error by using BigDecimal for example in java, the largest fraction is 1000000000 / 999999999 = 1.000000001000000001000000001, the numbers after the 1. well represent the rounding we will use inside the BigDecimal class, 51105537

»
5 лет назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

In E. Why can't we just simply use two pointers. My idea is to sort this array then use two pointers as below a[n+1] = 2e9; for(int l = 1, r = 1; l <= r && r <= n+1;) { if(a[r] — a[l] <= 5) r++; else{ len[m++] = r-l; l = r; } } sort(len, len+m, greater()); for(int i = 0; i < k; ++i) ans += len[i]; But I get wrong answer on test 25. Can some one help me?

»
5 лет назад, # |
Rev. 7   Проголосовать: нравится 0 Проголосовать: не нравится

DP in editorial for E is something different from its definition.

In that solution we just skip intermediate students (i, i + cnt_i), but in this case dp[i + z][j + 1] (z = 1 ... cnt_i — 1) is not increased as it should be.

Consider following example: n=3, k=1 a[] = 1 1 1

Then editorial code outputs: dp[i][j]=

0 0 
0 0 
0 0 
0 3

But by definition dp[1][1] = dp[2][1] = 1, because we form one team from these students.

Can someone please explain what dp[i][j] really stands for?

»
5 лет назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

can E be solved using divide and conquer after sorting?

»
5 лет назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

Can anybody tell me why this solution got TLE my submission

»
5 лет назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

Can someone explain why the maximum number of students is possible only when there are exactly k teams as written in the solution for E?

»
5 лет назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

f2 is missing!

»
5 лет назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

my solution for F1 https://codeforces.com/contest/1133/submission/51458164 is giving correct output on my computer but I am getting wrong answer on codeforces. I am not able to understand the reason. Can anyone help me out ?

  • »
    »
    5 лет назад, # ^ |
      Проголосовать: нравится 0 Проголосовать: не нравится

    I saw your code and ran it on ideone. It works fine on ideone. So I submitted it just to be sure and got WA, just as you got. Then I noticed the size of the vector you allocated in the bfs function. It should be n+1. I got AC after changing it. https://codeforces.com/contest/1133/submission/51472660 This is the AC code. Only the size is changed, nothing else. It worked on your machine and ideone probably because it allowed out of bound call, and luckily there was no garbage value in that memory.

»
5 лет назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

In problem E, why can't we just loop k times, in each loop get the team with max members under the given condition, take the remaining students and continue till k teams are formed. My submission

  • »
    »
    4 года назад, # ^ |
      Проголосовать: нравится 0 Проголосовать: не нравится

    I was doing the same thing and got wrong on 52 as well, did you figure it out? Please tell me if you did.

»
5 лет назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

Can any one help why my binary search solution isn't working? https://codeforces.com/contest/1133/submission/55474924

  • »
    »
    5 лет назад, # ^ |
      Проголосовать: нравится 0 Проголосовать: не нравится

    When passing a vector to a function, you should pass it by reference, otherwise it may lead to TLE, just like it did for you (read more here). Also, in your binary search function, you do not define your variable "ans" when you create it, which can be problematic if you never set the value during your binary search.

    I would also suggest using standard C++ lower_bound or upper_bound functions when doing a binary search on a sorted array (or in your case, a vector). It might take some time and practice to learn how to use them but it will be really useful and you won't have to write your own binary searches when it is unnecessary.

    • »
      »
      »
      5 лет назад, # ^ |
        Проголосовать: нравится +3 Проголосовать: не нравится

      OMG, this silly mistake I was doing it whole time in many questions, thanks a lot for pointing out, just checked lower_bound function it's really helpful. all the three tips were helpful, BTW, can you help me with this dynamic programming question https://codeforces.com/contest/158/problem/E I am really stuck, not able to understand how to form the base cases (i,e. initial values of 2d array dp[][]). Thanks Again

      • »
        »
        »
        »
        5 лет назад, # ^ |
          Проголосовать: нравится 0 Проголосовать: не нравится

        I can see that the editorial of that contest (link) has a few comments regarding this problem and you might find them helpful.

      • »
        »
        »
        »
        5 лет назад, # ^ |
          Проголосовать: нравится 0 Проголосовать: не нравится

        Also, when you need help with a problem it is a good idea to ask for help by creating a separate blog for it, I have noticed that people tend to answer these more often than comments under blogs like this.

»
4 года назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

For 1133E — K Balanced Teams, I've used a backward DP that's my best friend, unlike the forward DP given in the editorial.

Here is my submission.

  • »
    »
    4 года назад, # ^ |
      Проголосовать: нравится 0 Проголосовать: не нравится

    Can you explain your solution?

    • »
      »
      »
      4 года назад, # ^ |
        Проголосовать: нравится 0 Проголосовать: не нравится

      It's pretty much like in the editorial. The $$$cnt_i$$$ in my solution represents the number of students in a team, if the $$$i^{th}$$$ student is the last one in it.

      And the dp goes in the similar manner. The $$$dp_{i,j}$$$ represents the maximum total number of students, in upto j teams, while considering the first $$$i$$$ students.

      You can decide to take or not take a student, and the dp simply states that, in the for loops.

      It's been 3 months so I had to look into the solution itself again to understand what I'd done. Do let me know if you need more clarification.

»
4 года назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

Can someone help me figure out why am I getting TLE. I have written almost the same code as in the solution provided.84302770

  • »
    »
    4 года назад, # ^ |
    Rev. 2   Проголосовать: нравится 0 Проголосовать: не нравится

    You can solve that problem C by map easily.

    Code
»
4 года назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

vovuh, editorials are not linked to contest materials of problems. can u please look into it?

»
4 года назад, # |
  Проголосовать: нравится 0 Проголосовать: не нравится

Why in solution he shuffles the output ?