Tutorial is loading...
Tutorial is loading...
Tutorial is loading...
Tutorial is loading...
Tutorial is loading...
Tutorial is loading...
Tutorial is loading...
№ | Пользователь | Рейтинг |
---|---|---|
1 | tourist | 3985 |
2 | jiangly | 3814 |
3 | jqdai0815 | 3682 |
4 | Benq | 3529 |
5 | orzdevinwang | 3526 |
6 | ksun48 | 3517 |
7 | Radewoosh | 3410 |
8 | hos.lyric | 3399 |
9 | ecnerwala | 3392 |
9 | Um_nik | 3392 |
Страны | Города | Организации | Всё → |
№ | Пользователь | Вклад |
---|---|---|
1 | cry | 169 |
2 | maomao90 | 162 |
2 | Um_nik | 162 |
4 | atcoder_official | 161 |
5 | djm03178 | 158 |
6 | -is-this-fft- | 157 |
7 | adamant | 155 |
8 | awoo | 154 |
8 | Dominater069 | 154 |
10 | luogu_official | 150 |
Название |
---|
Good!
Why?
For problem Div2 E, 99886275 is just so cool.
Thank You.
Nice Problemset — problems were to the point and good.
I really liked it.
I used dictionary tree to passed the div2 D, I can guarantee its correctness, but I can't guarantee its time complexity,can anyone hack my solutions? https://mirror.codeforces.com/contest/1457/submission/99900717
In Div2D, why is n<=60 if the given condition is true?
Because if n>60 , We can always find a contiguous triplet (a,b,c) such that a > b^c.
Why we can always find that?
In case $$$n > 60$$$ There will be three numbers $$$(a, b, c)$$$ all having the same msb (Most Significant Bit) set to 1. You can do one operation with numbers $$$b, c$$$ this will turn off the msb therefore $$$ a > (b \oplus c)$$$.
Thanks
Because a_i <= 10^9 ~ 2^30. So of numbers which are k digits long in binary, we can have at most 2 to not get a triplet. And there are only 30 different sizes before we hit the cap of 10^9.
edit: I may have answered the wrong question
Thanks
Because these three numbers share the same highest bit $$$1$$$. And $$$1 \oplus 1=0$$$. Let's consider these three numbers as $$$a,b,c$$$ where $$$a<b<c$$$ , then the highest bit will become $$$0$$$ after we do $$$b \oplus c$$$, so $$$a > b \oplus c$$$ holds. Then we can always find that.
Thanks
Me when doing normal rounds: So I will try to do problems in order, if I can't do A in 30 minutes then I will consider B...
Me when doing russian rounds:
For Problem B, n is <=10^5, test cases are <=10^4 and no. of colours are <=100. How is 10^11 (that's what I can interpret) solution works??
If some one has any explanation and would like to correct where I am wrong, it will be great :)
It is guaranteed that the sum of n over all test cases does not exceed 10^5.
So when value of n is 10^5 then no. of test cases will definitely be 1. When test cases are 10^4, then value of n in all test cases are such that their sum will be atmost 10^5.
There is a statement, "Sum of $$$n$$$ over all test cases doesn't exeeds $$$10^5$$$". That basically means $$$t*n<=10^5$$$. So, overall operations doesn't exeed $$$100*10^5=10^7$$$ which is acceptable in the given time limit.
I did not see that B had only 100 colors :/
I am still confused in Div 2D.
plzz anyone can explain div2 D. why n<=60
For n <= 1e9, There can be atmost two numbers having same most significant bit. Therefore the bound will be atmost twice the number of bits i.e. 2 * log2(1e9) which is 60.
thanx bro!
Thanks for detailed explanation of Div2E!
For Div 2 Problem A, this is my code
I have calculated the Manhattan Distance of the cell (r, c) from each of the corner cells, and printed the max distance. My code is giving wrong answer on test 2. Can anyone tell what's wrong in my code?
abs(a — b) + abs(c — d) is not equal to abs((a+c) — (b+d)).They both are different.
find abs of (row) and col seperate for all four corner
int ans=Math.max(row-1+col-1, Math.abs(n-row)+Math.abs(m-col));
for D, I brute forced subarrays of size <= 31, then brute forced again on the "split point". for example, if i have a subarray i...j and a split point of k, we xor the range from i...k and k+1..j. then, we check if the sequence we made decreases at any point. why does this work? I used the claim that there does not exist a construction where we have to make > 29 moves.
https://mirror.codeforces.com/contest/1457/submission/99871104
Nobody found the $$$O(N)$$$ for div1D?
how to solve div1D in $$$O(N)$$$ ?
For each situation where you just took the $$$i$$$-th thing, find the set of positions of the clone. For each situation where the clone just took the $$$i$$$-th thing, find the set of your positions. In the first case, it's a contiguous range, which means it has to be up to 2 contiguous ranges in the second case. There's a lot of possible transitions from $$$i$$$ to $$$i+1$$$ but it's a constant number and each takes constant time.
Can anybody explain the prefix suffix method used in division 2 in D-problem?
Is it only me or Div1 E editorial is cryptic.
"Let call low bits that we don't need to care free bits." — can you maybe give an example.
What is A there?
For example, assume you have segment $$$(10, 20)$$$ and you set your number $$$011??$$$, your number will be in the segment whatever you set two lowest bits, then they are free.
in div2C,why the answer of 2 2 1 10 11 1 is 11,if i make the 0 to 1,it just take 1 second
This is your test case:
n = 2, p = 2, k = 1; s = "10"; x = 11, y = 1.
x is the cost of adding a platform, y is the cost of removing the first cell. In your case, there is only 1 option: that is to add a platform at position 2, giving 11 as your answer.
KAN For div2 D, the editorial states the following:
Indeed, there are no two integers with the same highest bit set. It is much easier to solve the problem in such constraints.
I think it should be stated as: there are at most two integers with the same highest bit set. Right? Because for input [1, 6, 7], the highest set bit of all numbers are different but 6 and 7 have the same highest set bit.
Ya! Exactly. I too think it's a typo.
Thanks, corrected! I missed "more than".
How is the space complexity for div2B O(N) ?
You are essentially keeping a separate index list for each possible color. The total houses is N, so the total entries in all index lists sum up to N.
It is good tutorial.
Kindly do mention the code also. Sometime explanation is not sufficient
Could someone please explain the dp approach to div2 C?
https://mirror.codeforces.com/blog/entry/85081#comment-727466
I will explain more the 4 and 5 hints. - Given that the first cell on which the ball will land is cell i, we know it will visit cell i, i+k, i+2k, ... , i+xk<n, right! so we want to count how many index j in {i, i+k, i+2k, ..., i+xk<n} such that s[i] = '0'. In other words, we need to how many cell in the group of cells the ball will visit, does not have a platform. we will of course need to put a platform in each one of them. we have no choice.
Let cnt[i] denote the min number of platforms we must add if the ball starts the journey at index i.
Let's consider the bruteforce first. It's a simple for loop.
we simply iterate over indecies i, i+k, i+2k, ... and if no platform we increment a counter. right!
Now, we need to optimize this code using DP
Thank you!
It seems that the standard program to the Div.1 D has something wrong
https://codeforces.ml/contest/1456/hacks
Fixed, some testers' solution got hacked too.
What would be the difficulty level of XOR gun?
I bet 1800
Nice trick in problem Div2D =)
Sorry for my stupid, it's a little hard for me to understand Div1E solution, can someone explain it easily? thanks!
1415B - Repainting Street
Can somebody explain why test case 3 in test 2 (below) says right answer is 2 days? If we choose any color (e.g. 1), there are only 6 houses, which are not this color. So, with capacity k=6 one day should be enough to paint. Looks I am missing something, but don't get what.
$$$K$$$ means a range that must be consecutive.
So in the test,you can choose [1,6] or [2,7] or [3,8] or [4,9] to change the color.
Hey guys, here is my video proof for "Genius Greedy" in Problem E using induction hypothesis. Anyone who is interested can take a look