maroonrk's blog

AtCoder Regular Contest 133 Announcement
By maroonrk, history, 17 months ago, In English

We will hold AtCoder Regular Contest 133.

The point values will be 300-500-500-700-800-1100.

We are looking forward to your participation!

  • Vote: I like it
  • +215
  • Vote: I do not like it

Comments Comments (13)
| Write comment?
»
17 months ago, # |
  Vote: I like it +56 Vote: I do not like it

Another maroonrk round!

→ Reply
»
16 months ago, # |
  Vote: I like it 0 Vote: I do not like it

someone who did Problem B using bipartite matching ?

→ Reply
  • »
    »
    16 months ago, # ^ |
    ← Rev. 2   Vote: I like it 0 Vote: I do not like it

    Is there a solution using bipartite matching?

    UPD: the answer is no, why downvotes?

    → Reply
    • »
      »
      »
      16 months ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      I was thinking of doing something using Hopcroft but was not able to implement, asking here if somehow someone was able to figure out how to cancel out alternating edges to find maximal matching

      → Reply
»
16 months ago, # |
  Vote: I like it +10 Vote: I do not like it

Problem B reduces to a previous atcoder problem called "cross-free matching" https://atcoder.jp/contests/arc126/tasks/arc126_b

→ Reply
»
16 months ago, # |
  Vote: I like it +6 Vote: I do not like it

Can anyone explain why we sorting as (i,-j) in B.

→ Reply
  • »
    »
    16 months ago, # ^ |
    ← Rev. 2   Vote: I like it +3 Vote: I do not like it

    Consider another defintion of the dp:

    If we go from left to right in p[], we want to pair the current element p[i] with one of the q[j]. We previously searched all indexes in q[] where we find multiples of p[i], let pos[i] be the list of indexes in q[] that are multiples of p[i]

    So, let dp[k]=max possible length of an common subsequence ending in q[k]. Then

    dp[pos[i][j]]=max(dp[0..pos[i][j]-1])+1 for all positions j and current i

    We can solve this with an extra structure like a segment tree. Or, we just sort all the indexes in pos[i] biggest first, and create the LIS on that list.

    Edit: Changed LCS to LIS

    → Reply
    • »
      »
      »
      16 months ago, # ^ |
        Vote: I like it +3 Vote: I do not like it

      What is LCS referred to? Can you explain by taking an example.

      → Reply
      • »
        »
        »
        »
        16 months ago, # ^ |
          Vote: I like it 0 Vote: I do not like it

        Actually I wanted to write LIS here, as mentioned in the tutorial, the longest increasing subsequence.

        I think I mixed it because the problem is related to the LCS, the longest common subsequence, see LCS

        → Reply
»
16 months ago, # |
  Vote: I like it 0 Vote: I do not like it

I have a different approach for C.

If we write the sum as $$$\sum_{i=1}^nA_i+K*w_i$$$. I claim that $$$w_1,w_2,...w_{n-1}$$$ must be maximized, we can use adjustment to prove it. Under this constraint, we want to maximize $$$\sum_{j=1}^{m-1}a_{n,j}$$$, this is a simple greedy problem.

→ Reply
»
16 months ago, # |
  Vote: I like it 0 Vote: I do not like it

For problem F, how to compute $$$w$$$ that mentioned in the editorial?

→ Reply
  • »
    »
    16 months ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    Now I know this. But I'm still confuse about how to compute it by divide and conquer. His code is too hard to understand.

    → Reply