You can ask me anything and I will try my best to answer. As a problem author of a successful round and an Master I think I am qualified to do AMA.
# | User | Rating |
---|---|---|
1 | tourist | 3947 |
2 | jiangly | 3734 |
3 | Radewoosh | 3646 |
4 | jqdai0815 | 3620 |
4 | Benq | 3620 |
6 | orzdevinwang | 3612 |
7 | ecnerwala | 3581 |
8 | Geothermal | 3569 |
8 | cnnfls_csy | 3569 |
10 | ksun48 | 3479 |
# | User | Contrib. |
---|---|---|
1 | awoo | 162 |
2 | maomao90 | 160 |
3 | nor | 156 |
3 | adamant | 156 |
5 | -is-this-fft- | 155 |
5 | atcoder_official | 155 |
7 | maroonrk | 153 |
7 | cry | 153 |
9 | SecondThread | 147 |
10 | Petr | 146 |
You can ask me anything and I will try my best to answer. As a problem author of a successful round and an Master I think I am qualified to do AMA.
Name |
---|
What do you think about developing logic for AD-HOC problems ? do we need to prepare for it diffrent then other type's of problems ?
You need intuition in those problems.
To develop intuition, you have to practice. Record your thought process. If you can do the problem, take a look back what steps have you taken to get you there. If you cannot do the problem, compare what you've got and the editorial, and see what you are missing, Repeating this a lot of times will help you to know what to do soon after you see a problem.
What's your idea in this problem
Basically I would try out small cases, and see how well it generalizes. Then it would be some constant optimizations on the number of vertices needed.
I have not actually done the constant optimization part yet tho. I think it needs some clever way to optimize the number of vertices.
How can I get comfortable in constructive problems and come up with solution faster most of times I don't even come up with solution ?every constructive problem is different.
Literally just practice this is the 30520th question asking this. There is no shortcut. Practice
I agree that practice is the most important. The most important thing to train is your intuition and your pattern seeking abilities.
In a contest, you should try out small cases and see if you can spot some pattern that can be generalized. This is how you use your intuition and pattern seeking skill.
Sometimes the pattern you tried to conclude is wrong. Then try again. Constructive problems have a lot of trial and error.
Who are yo u ? ??? ? ?
he's the local farter in my locality he's a big bully he keeps bullying me i feel very sad every time i see him cuz i know he will grab me and throw me into a trash can :sob:
Which means he is a CHAD
OMG SPARKY MASTER IS BACK, IS ALIVE!
MASTER PLEASE TEACH US YOUR WAY
Notice me Sparky_Master_WCH1226
I am a 15 year old Hong Kong student.
Not again!
how so good at reaper
also opinions on [REDACTED]
Can you first teach me how to get near perfect score in IMO on your first year?
Solve 6 problems and get docked 1 point
if yes who u main?
Sadly no
hey how to approach Tree Dp problem like rerooting and dsu on tree
Actually I do not think a newbie should learn tree dp yet. You should consolidate your basics first.
But anyways, dp is a tool to use when there are a lot of overlapping sub-problems. You should use the fact that the number of edges of a tree is linear and that there is no cycle to help the dp.
Im newbie on cf only :) otherwise i have solved of lots of Data structure and algorith problems and Leetcode now i am trying cp but its quite hard not able to solve qns thats why i ask
How to boost my cp??
Just practice. Since your question is quite generic I cannot give you a concrete answer. You can scroll through other comments of mine here on this post to see some more advice.
But how practice......like given virtual contest or selecting problemset and apply filter upto level i.e like 1200-1400.......or any other else......Please give me one solution......i am little confused.
What is the most interesting problem you have solved (in your recent memory)?
1874D - Jellyfish and Miku
when are you planning to retire?
Not anytime soon
CP is so fun :)
What is your favorite problem out of those that you made?
Favorite OI problem that is public already? 1815E - Bosco and Particle
Favorite OI problem that is not yet public? I have one that I really like, will definitely use for my next round.
Favorite math problem? The following problem has a really elegant and elementary solution, you can try it out!
Let $$$ABCD$$$ be a rectangle. Let $$$E$$$ be on side $$$\overline{AB}$$$ such that $$$BE=6$$$, and let $$$F$$$ be the intersection of $$$\overline{AC}$$$ and $$$\overline{DE}$$$. If the distance from $$$F$$$ to side $$$\overline{AD}$$$ is $$$15$$$, and the absolute difference between the areas of $$$\triangle AEF$$$ and $$$\triangle CDF$$$ is $$$40$$$, what is the area of $$$BCFE?$$$
hi bro, would the area of BCFE be 140? seems like an interesting puzzle!
Yes you are correct!
yay nice thanks
-what is the best way (from your opinion) to study&practice DSA? should i study them first then start solving random problems? or should i solve first and when i face a new concept i go and learn it?
-also if you can recommend a good source (videos/books) for studying DSA? and if i'm facing a new topic (let's say binary search) how much problems tagged with it should i solve? how to choose them?
-also how much DSA i need to reach specialist?
-lastly if you have any general advice that will have a good impact in CP it will be appreciated (things like how to stay focused, deal stress while doing contests or whatever you see helpful..) thanks in advance :")
I think you should learn all the things you need for your skill level, solving a few example problems for each thing you learn. Then you solve problems of your rating to your rating plus 300. Do that until you can confidently solve those problems before moving on to new topics.
I think that Codeforces has a lot of good blogs on different DSA. Just search them up. I think it would not be hard to distinguish a high quality tutorial with a low quality one (you can look for the rating of the author). There will be some recommendations on what problems to solve on those blogs. The amount of problems you solve is not fixed, as I have said in the previous paragraph, solve until you feel confident in solving problems of a similar difficulty and the same trick in the future. In addition, if you have a coach, you can also ask him/her for assistance on what tricks to learn and what problems to solve.
You need to know !D and 2D partial sum and difference array, two pointers, binary search, and the built in data structures (stack, queue, priority_queue, deque, vector, map, set, multiset if you are using C++) and math stuff to reach specialist. Actually you only need those until Candidate Master.
I think an advice is that if you think you are exhausted, try taking a break from CP. This can help you recover, which will make you better.
Next is how to deal with contest stress (what you asked for). In a codeforces contest, if you are not doing well, just remind yourself that it is just a codeforces contest and you should not be too stressed about it. In a cp contest that actually matters, you should try to see how can you get the maximal score in the remaining time you have. If you are really stressed out drink some water and go to toilet to calm down.
That is all that I have!
thanks a lot for all of this! only one more question, regarding "math stuff" that you have mentioned, what exactly do i need to know?
Combinatorics and some equations involving floor, mod and bitwise operators. Look them up!
okay, thanks again <3
when should i solve problems and when should i learn new algo? and when it's very hard for u to understand smthg in the learning site (like a book), what would u do?
You do not need to know any algorithms and data structures except
You have to be good at making observations, and sometimes deriving formulas. That is why I recommend you to learn basic combinatorics and some identities about mod, floor, and bitwise operations.
So, just learn the above and you can start doing problems until you are confident to do all problems below expert. Then you can learn more advanced algorithms such as dp and algorithms about graph, and data structures like sparse table and segment tree. But bear in mind that you can reach Candidate Master with only the basic stuff that I mentioned.
Also when I read a book and encounter something that I do not understand, I will try to skip it first and read stuff that is not dependent on that thing I do not understand.
do you think i should learn all the basic and master them or learning new algo ? i'm having a competition 3 months later in my province, im pretty sure that I will get a high score, but only in my province, when I go to the national OI, I can see that im terrible. Is that any way to learn cp fast(like reaching M in 1 year)?, Im 13 and only started cp since the beginning of 2023 (most of the kids who good at cp in my nation started since they were around 8). I really love to try IOI tho.
I think you should learn all the basic first.
If you want to reach Master in a year, you should spend more effort, and also I recommend you find a good person to coach you and support you.
Also, I think Master is enough for IOI in most countries (except the top countries in which you might need to be IM or even GM).
how can i know what algo i need? and how to learn an algo efficiently ? when learning a new algo, I just go to an online website and read, then at the end, I just solve some basic problems they recommend, but when I try a real problem which need that algo, it's like a whole new level
I really recommend you to find someone to teach and guide you. My school's coach helped me and my schoolmates a lot. Now in my school, within those currently studying, four of us are Masters or IM.
what do you do when you feel tired or bored when doing cp? also, it's little bit hard for me to find a good teacher in my province ( pp don't care much about cp), nearly 0 middle school teacher can do cp good in my area lol, and even some teachers in high school, they just teach too slow, they tend to teach a whole class instead of just some people.
I either do schoolwork or maths when bored doing cp. Also I think you can find somebody who just went to university and good at cp, my coach is doing phd now and he is quite free so he can be our school's coach.
is hong hong a country or province of china?
Hong Kong is a special administrative region of China
Bruh! Looks like you hibernated after writing the blog.
No I am just late to reply
How often do you practice cf , excluding contests
You might see I rarely submit on codeforces except for contests.
This is because there is a online judge only for Hong Kong students and I usually do the problems there. Also, when I do problems on codeforces, I do the problems from mashups set by my school’s coach so you won’t be able to see it.
Have you participated in IOI?
No, but I have participated in IMO 2023.
Do you think genius is more important than practice in competitive programming?
Would your answer be the same for mathematics olympiad ?
I believe that gift will affect how quick you can learn, but a gifted person must still learn and practice in order to become better. This applies in literally anything.
Do you love gKseni too? :D
I LOVE her very much. :kisses:
Sorry but I don’t know who’s gKaeni
How important is math for cp
Very important. Basically, the problems that you are able to solve is limited by your math ability.
Even in data structure problems, you have to come up with formulas. If you cannot do algebra correctly, you cannot solve the problem.
I dont have questions for you. Actually, you can ask something from me, but I need some time to think.
I have finished answering above questions.
culver0412, what is bruh open?
I think arvindf232 is a better person to answer this question.
bro doesn't even know time complexity of fft
What's your main judge for practicing? What other judges you use?
Where can i start if I'd like to build a good math base?
Can anyone be a red?