What is your least favorite type of problem / least favorite topic? I personally do not like graph problems.
# | User | Rating |
---|---|---|
1 | tourist | 4009 |
2 | jiangly | 3823 |
3 | Benq | 3738 |
4 | Radewoosh | 3633 |
5 | jqdai0815 | 3620 |
6 | orzdevinwang | 3529 |
7 | ecnerwala | 3446 |
8 | Um_nik | 3396 |
9 | ksun48 | 3390 |
10 | gamegame | 3386 |
# | User | Contrib. |
---|---|---|
1 | cry | 165 |
2 | maomao90 | 164 |
3 | Um_nik | 163 |
4 | atcoder_official | 161 |
5 | adamant | 160 |
6 | -is-this-fft- | 158 |
7 | awoo | 157 |
8 | TheScrasse | 154 |
9 | nor | 153 |
9 | Dominater069 | 153 |
What is your least favorite type of problem / least favorite topic? I personally do not like graph problems.
Name |
---|
Constructive. If every problem was constructive, i'd probably be a pupil
Same. I would say constructive too, but it becomes my favorite problem type after I manage to solve one by myself.
Can you define what a "constructive problem" is to you? The people that I've talked to about this don't have nearly the same opinion as I do.
Questions like these where you just have to think of a pattern
https://mirror.codeforces.com/contest/1930/problem/B
https://www.codechef.com/problems/STACKED
I agree that it's fair to call those problems constructive. For the first one you can simply try all the common answers for the problem and one of them works.
1 2 3 4 5 ...
1 N 2 N-2 3 N-3 ...
N-1 N-2 N-3 ...
etc
The main observation being that the smallest multiple of x that isn't x is 2x. Trivial observations are easy to miss and I didn't think of that until finding the construction.
The second seems to be a mix of greedy thinking (use big numbers to escape the sum range when you get to it) and a well known greedy problem (with numbers [1,x] you can always find a subset of sum for every value in [0, 1+2+3+...+x]). I've seen easier problems in div1E.
I would describe as "problems in which you're given a set of properties and aims to find a structure that matches them"
I don't agree. It depends on the solution. For example: if the solution is some DP then DP is the main part of the problem.
Same man. It is like a luck thing. Either you get the pattern or not.
Gotta confess, skill issue for me. And I don't even wanna upsolve/practice such problems
There more general ideas that motivate constructives than you think once you start reflecting more and put extra effort in trying different approaches. This is coming from someone that is specifically solving more constructives to do better in contests. Sadly something that is necessary to improve on (with pain in the process), but I think being more open minded on this type of problem helped my solving.
True that. I have realized that I should stop whining and get better. Will practice such problems more, hopefully. Thanks a lot for the advice! :)
It is more of an IQ thing than a luck thing.
I get where you are coming from, but I have a well respected teacher, who says that a person who can solve a 3500 rated dp/ds but can't do 3400 constructives, you're doing something wrong.
[This is a personal opinion from me, not my teacher]: Constructives force you don't think to solve the problem in known or standard ways, rather forces you to think differently and out of the box; yes it can be frustrating but when you're trying a constructive problem, your brain thinks of 1000 different new/cool/weird ideas + observations + etc which can come very helpful in the future problems, and it also boosts your creativity
"which can come very helpful in the future problems"
Untrue
for some constructive yes, this may be untrue, but there are good ones out there that will make sure to make this statement be true
Yea your teacher spitting fax. I probably hate them because of my skill issues
I used to dislike constructive problems too, but from a certain point on, I just started enjoying how most of them rely on simple patterns.
I am also very weak and get stuck on every easy constructive problems. But I still like them most.
Bits
+
1995C - Squaring
1995B2 - Bouquet (Hard Version)
these are worst type of problem that I could imagine.
specifically problem using formula/math problem, like translate problem statement and then answer be like one line max(0, c[i-1] + ceil(log2(log2(a[i-1])/log2(a[i])))) * (k1-r)*x+(k2+r)*(x+1) where k1=min(c_x,m/x) and k2=min(c_{x+1},c/(x+1)) .. it's fucking insane actually
Nah, maybe the formula should actually be:
272232560
why are you multiplying it with (k1-r)*x+(k2+r)*(x+1)? I believe there's no need.
It's not hard it's just math.
Graph, data structures and implementation
I feel like implementation-heavy ones are fine unless you gotta implement them in a language other than python.
I dislike them for lacking some brilliant ideas(for the most part) and just being about writing a bunch of code
I feel like implementation-heavy ones are fine unless you gotta implement them in a language other than C++.
don't get me wrong, I've done python for 2 whole years, but not for cp, for projects, my whole github is python-based, but for cp I'd rather do C++ A hundred times, You know, if you learn it you'll love it, the problem is everywhere u go they try to teach the things that are useful in projects, never in cp; but learning C++ for cp is easy and almost super fast(nearly my whole class did in like a week at max), so yeah C++ is the one for cp, at least for me
I don't like constructive ones
hard problems, but seriously, heavy implementation
I don't usually hate problems, I hate badly-written tutorials.
If a tutorial confuses you, it's surely the author's fault. - PinkieRabbit
I hate bitmask problems.
implementation
Problems with no thinking or brute forcing and noticing a pattern which 99% of participants can't prove.
exactly, like this one 1983B - Corner Twist
What's wrong with it, I meant something like this 1916D - Mathematical Problem.
I hate this problem too.Regarding 1983B - Corner Twist,it needs you to figure out that the row-wise and col-wise sum % 3 remains constant,which as such might not be hard to see. But it's certainly not easy to prove that this is sufficient as well(atleast for me).
Well I think everyone just has different strengths, because it was quite apparent to me from the beginning that the condition is sufficient.
this is just bcuz u're a noob, this is so easy and clear to prove that it is necessary, and to prove it's sufficient you just say: I can merge 2 operations done on a single vertex
skill issue ig :(
Geometry.
Geometry + graphs.
https://mirror.codeforces.com/blog/entry/78046?#comment-631220
yeah geomentry sucks
interactive...
Interactives are fun
Constructive and matrix implementation stuff
combinatorics and number theory
combinatorics is one of the cooler ones out there imo
now that i think about this i only hate problems that are hard to implement i really never mind the topic i sometimes even solve geometry ...
number theory
DP: specifically plug, digit, bitmasks, and those where you have a complicated state and transition which often involve math.
IQ problems
bitset
then why have that user name?
*sarcasm
Bit manipulation
Ad-hoc
Graph, problems with segment-tree/sqrt-decomposition and problems with fractional numbers.
Edit: And yeah, the most important. Games.
orz Implementation eric mathforces
flows and undirected graphs
Problems that are stolen from other sources
problems that i can't solve
Constructive/ad-hoc/guessing problems. Or when the proof is harder than the problem.
Ad-hoc problems in general.
Especially problems where you are given an array (or a binary/ternary/etc. string) and are asked to find the minimum number of operations to make the array satisfy a certain condition.
Agree with you, sometime I think I was the most stupid person in the world because of not solving that problems :(((, maybe every problems like that
Data Structures.
When I'm coding for myself I love DS. When I'm coding for a contest I hate DS.
Graph problems really piss me off.
Bitmask and Interactive
Constructive problems first of all
Guess problems ( I mean not all of them are bad. If some observations lead you to the guess then this type of problem is fine. )
And math problems that are more suited for IMO or national mathematical rounds and you need to have knowledge of some theory that barely anyone knows.
And finally single trick pony problems (If you know it then you can solve it in 5 minutes. Otherwise good luck)
:v
Yeah I one time saw a problem that required knowledge of "the toilet function" or something. They're really running out of names for these arcane math things.
Do u mean euler totient function?
yeah that was it actually
Greedy+Games. With my slow brain and deterministic mindset combined, it's really hard to come up with an idea, assure that it would be correct, and keep that assurance solid enough so my mind could force itself to write the code. Or I'm just bad in heuristics-based stuff. :<
Problems that have strings or game theory. I always get super confused thinking about cases in Alice Bob problems.
interact problem
so there is a catch with interactive problems. solve one of them and you'll love to solve the next one.
bro you are right ,now I can solve some ez interactive problems, it is so interesting XD
Expert soon...
Anything that doesn't even slightly resemble a real world process.
I Hate Extensive and Long Implementation Problems :)
constructive problems , especially the ones which you need to generate a permutation which satisfies a random constraint
Casework problems.
Just.. ewww.
Those problems whose logic is pretty simple but have annoying edge-cases. Simple problems with many cases to consider and there is no pattern to use loop.
Geometry + Constructive + implementation
Constructives are the reason of me tanking the contest at which they appear especially at C.
brute force and greedy.
Pretty Interesting!
Greedy
Games and number theory
Geometry is shit
Dp Is Shit
Case work is Shit
DP. I hate DP :(
Constructive, combinatorics and very gross math problems
Combinatorics is the best tag IMO
long code problems
ad-hoc stuff where 90% of the time the solution is guessable
I think the problems I like the least are problems that have way too much emphasis put on knowing some (relatively) hard algorithm or data structure, and otherwise the problem is pretty easy. This is even worse when such problems appear in onsite contests without internet access.
when you need to use some linear things on a graph or a tree.(for example:segment tree on graph)
geometry
games ofc, I cannot do optimally any case
dp+graphs... i love dp ...but when it gets merged with graphs i just hate it
you should get this shirt
I hate sigma problems.
Problems that i can't solve like C<D<E<F<G<H they take my entire time just to read and understand question then implement correctly, sometimes i think solving problems more than anyone did will make me better than them cause u will be familiar with ideas and implementation and you u dont have to waste your time reading, solving, implementing, debugging
String, heavy DS, heavy math
Math or data structures:(
Probability
for me its number theory
dp
1) Implementation heavy problem. (probably too many corner case are not my type.)
2) Every problem that require reconstruct the operations / rebuild the graph, rather than just asking number of operation.
may be, i am just saying all kind of problems i am weak at/
I would probably say the problems I struggle solving, which would probably mostly fall in implementation, and bitmasks. Bitmasks mostly because I am not too familiar with the topic and am still learning it, and implementation mostly because my program will fail some edge cases, and when solving such problems I always feel like diving into more particular cases and not getting a general solution. That is why, even when getting the right solution something feels wrong with implementation problems.
A more positive question I would have is what is your favorite topic ? I would say mine are number theory (I have more of a mathematical background than Computer science one) and maybe two pointers.
i like tree, actually i like everything but i m just stupid
Data Structures , Math
graph
Math problems, especially where solution time complexity is less than O(N), they are usually hard or you just have to know some specific math to solve them
proof by AC problems.
least favorite type of problem: easy problems which should be solved, but was unable to solve during contest..
graph theory by far
Observations based problems
ad-hoc, there could be so non-straightforward things
idk if I'm just stupid, but I've only seen 1 straightforward codeforces problem in my entire life
They've put standards pretty high like TC — that's tough optimization stage, you may find out that HackerEarth on the early stages provided tasks where you have no clue what is required
interactive.How can I solve problems better than a robot?
bit manipulation lol
eeeeeeeeeeeeeeeeeeeeeasy to come up with the solution but diiiiiiiiiiiiiiiiiiiiiiiiiiiiifficult to write the code
Ad-hoc
I hate -... .. — .-- .. ... . / .- -. -.. / --. .-. .- .--. .... / — .... . — .-. -.--
I don't like geometry.
math and DP