Beknur-Tumen-045-26's blog

By Beknur-Tumen-045-26, history, 2 months ago, In English

Hi everyone! It has been 2 years I can't reach pupil in cf. If you have some experience from it help me please and give some problems to practice.

Sorry for my English.

»
2 months ago, # |
  Vote: I like it 0 Vote: I do not like it

Auto comment: topic has been updated by Beknur-Tumen-045-26 (previous revision, new revision, compare).

»
2 months ago, # |
  Vote: I like it 0 Vote: I do not like it

don't solve 800 problems only. You've solved 150+ 800 problems and comparatively, none of other difficulties.

»
2 months ago, # |
  Vote: I like it +5 Vote: I do not like it

Just my suggestion . Which is kind of working for me. Dont miss any contest. Give all the contests if possible and learn

»
2 months ago, # |
  Vote: I like it 0 Vote: I do not like it

For the past month your problems avg. rate is 900, solve harder problems.

Also checkout EDU courses, they will help you learn base structures and algorithms.

Don't jump in Suf Array/Seg Tree/etc.. Start from binary search/two pointer/prefix sums/etc. There are plenty of posts with great problems and topic lists.

»
2 months ago, # |
  Vote: I like it 0 Vote: I do not like it

I recommend solving CSES problems (at least 5-6 in each category). If and only if you can’t come up with a solution after 3-4 hours, you should google it.

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

    3-4 hours is too much i guess.

    • »
      »
      »
      2 months ago, # ^ |
        Vote: I like it +8 Vote: I do not like it

      I believe that the longer you think about a problem, the more skill you will gain from solving it. And I mean really think about the problem, go through all possible ideas, and not just wait for a while and then look at the editorial. I agree that maybe 4 hours is too much. You should think about the problem until you have no ideas left, for me this is usually 3-4 hours.

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

    the question is which 6?

»
2 months ago, # |
  Vote: I like it 0 Vote: I do not like it

try to solve problems at your limit: not to easy, not to hard.

learn new topics: learn about graphs, popular data structures, anything. don't know where to start? just google "competitive programming newbie topics" or smth.

struggling with a problem for 2 hours? read the editorial, think about what you have missed and try writing the solution again. still don't understand it? maybe its too hard for you at the moment, come back to it later

»
2 months ago, # |
  Vote: I like it 0 Vote: I do not like it

solve problems with rating higher than 1200. If you feel comfortable then increase the difficulty. Also make sure to upsolve contest problems.

»
2 months ago, # |
Rev. 2   Vote: I like it 0 Vote: I do not like it

You can try 1) CSES Problems and 2) TLE Eliminators cp-31 sheet(https://www.tle-eliminators.com/cp-sheet)

»
2 months ago, # |
Rev. 2   Vote: I like it 0 Vote: I do not like it

To become pupil you have to at least solve 1200+ rated problems consistently in practice. Do learn advanced algorithms right now like dp, segment trees, Dijkstra, etc. Learn binary search, prefix sums, number theory. If this is not enough, solve 100 problems of 800, 900, 1000, 1100, 1200. And you will reach pupil.

  • »
    »
    2 months ago, # ^ |
      Vote: I like it +8 Vote: I do not like it

    imo its clearly a waste to solve 100 problems of 800-900 difficults. most of them are pure intuition-based brute-force questions. you learn nothing. solving 1200+ questions only while practicing

»
2 months ago, # |
  Vote: I like it 0 Vote: I do not like it

solve from 1000 to 1300 rated 40-50 problems and give virtual contests... you'll reach there easily

»
2 months ago, # |
  Vote: I like it 0 Vote: I do not like it

To reach pupil you need these

  • you should be good at coding brute force solution fast

  • you should be able to code simulation fast

  • you know basic maths (like some modular arithmetic, GCD etc.)

You are currently 800, Solve 1000s and 1100s also start solving atcoder abc contest's first 2-3 questions

you can find the problems here: https://kenkoooo.com/atcoder#/table/

don't jump to the solutions keep trying, its okay to even try for days as long as you are able to think somewhere. only go to the solution when you are not able to think anything.

»
2 months ago, # |
  Vote: I like it 0 Vote: I do not like it

Solve as many as possible hard problems. (Hard means ~ your rating +300)

»
2 months ago, # |
Rev. 3   Vote: I like it 0 Vote: I do not like it

I had the same problem a few months ago. I couldn't solve at least 2 problems in Div. 2, 3-4 problems in Div. 3, and so on. What made me become a pupil/specialist was actually not grinding hard (my rating + 100/200) problems (but you should do that. Anyway). What I did was prepare for USACO. USACO itself is not easy. I could not solve at least 2 problems in USACO Bronze in 4 hours (which is how long the contest is). So, I solved problems at the USACO guide and learned new ideas on how to solve a problem, such as prefix sums. That's when I did so well that I was promoted to pupil. How I got into a specialist in a month is still a mystery to me, but I guess the best way to become promoted is to grind on hard problems (as before, you SHOULD try to solve hard problems) and potentially learn new ways of solving problems, instead of just using brute force all the time and getting TLE, for example.

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

    Which difficulty I should solve from USACO bronze?

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

      Slowly try to solve the easiest problem (which is usually the first problem) in each USACO Bronze contest (when practicing). Don't worry if it is too difficult as they are meant to be challenging. Try to solve bits and pieces of the problem, and maybe you can get all the test cases right! I would also suggest you do around 1000-1300 rated problems as the high-rated problems (1200-1300) can help you be lucky during contests while the slightly lower-rated problems (1000-1100) can help you consistently do well in contests. As your rating increases, so will the ratings of the problems you will have to solve.

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

      Also, if you also see Norp's blogs (80 days Pupil Challenge), he/she (don't know the gender, sorry) is solving 1000-1300 rated problems each day, which is great for him/her. If you can solve these problems, I bet you could become a pupil in around a month. It could take longer or less. If you can consistently do well in 1000-1200 rated problems, you can become a pupil in no time. In most Div. 2 contests, for instance, the first problem is usually an 800 or 900-rated problem while the second problem is usually a 1000-1200-rated problem. If you can solve those problems in a contest, you can easily become a pupil and if lucky, a specialist.

»
2 months ago, # |
  Vote: I like it 0 Vote: I do not like it

Less adhoc, more greedy. less 800 more >1000