Satwik_Mishra1's blog

By Satwik_Mishra1, history, 3 years ago, In English

This is to seek help from Experienced people on how to Improve,I have been regularly solving problrms from a long time now and I do try to appear In virtual contests almost daily. So is there some mistake in my practising strategy. If so please help me out. You may visit my profile for reference. thanks in advance.

  • Vote: I like it
  • -23
  • Vote: I do not like it

| Write comment?
»
3 years ago, # |
  Vote: I like it +33 Vote: I do not like it

I am not sure if I am qualified enough to write this but if you see your profile on cfvisualiser...

 . You will notice A steep drop from 800 to 900. So my guess is that you are solving problems that are too easy for you. You are doing hard work ,don't worry about the result. Sometimes it takes time to reflect your hardwork in your result!

  • »
    »
    3 years ago, # ^ |
      Vote: I like it -10 Vote: I do not like it

    Hey, have you looked at your own profile on cfvisualiser? I have the same "problem" and this can't be helped. Participating in many ranked contests naturally leads to having a lot of easy problems solved in our profile stats. Yes, it's possible to try to imitate rainboy and take care of problems in reverse order during ranked contests, but this is pretty unconventional. And will make our current ratings drop way below our current skills.

    Also having too many easy problems solved doesn't necessarily mean that too much time had been wasted on them. Easy problems are easy. Solving them is very fast. That high 800-difficulty bar on the histogram may be misleading.

    Doing virtual contests and ignoring their A & B problems rainboy-style to spend more time on higher difficulty ones may be an interesting experiment.

    • »
      »
      »
      3 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      Hey! I have seen my profile on cfvisualiser. Unfortunately this does-not happen to be the account I usually use for competitive programming. And yes I am guilty of having alt account but I was not aware of the rules earlier. On a side note it is not always true that people practice by doing virtual contest where they are able to solve only the first few problems. I usually practice by sorting problem set to be my current rating + 200 and solve those problems (on the other account obv). But it was a good suggestion! so Thanks for the same! Hope you have a great day and you calm yourself and be a bit less irritated!

      • »
        »
        »
        »
        3 years ago, # ^ |
          Vote: I like it 0 Vote: I do not like it

        I'm just saying that participating in ranked contests contributes to increasing the number of solved 800-difficulty problems regardless of whether we like that or not. The blog author participated in over 50 ranked contests and this surely contributes to having the same number of 800-difficulty problems solved in his profile stats (even if one somehow fails to solve problem A during a contest, I'm pretty sure that it's hard to resist the temptation to upsolve it later). Div2 virtual contests contribute to the growth of the number of solved 800-difficulty problems in the same way. And the blog author mentioned participating in a lot of virtual contests too. Having a high 800-difficulty bar on the histogram isn't surprising in these circumstances.

        Regarding your multiple accounts. I'm even not sure how is this supposed to be relevant to what we are discussing, But since you mentioned that, I'm curious about a few things. Are you scared to post comments from your main account for some reason? Do you feel nervous when participating in ranked contests to the extent that it affects your results in a negative way?

        I think that your way of cherry picking your current rating +200 problems from the problem set is also a good way to train. But it's important to have spoiler tags disabled in the account settings. Also I'm concerned that having no time limit pressure makes this kind of practice very different from real contests.

        Now I realised that skipping A & B problems in virtual Div2 contests isn't going to be very much different from just participating in virtual Div1 contests. I really want to give this a try and maybe the shape of my histogram will improve in the long run.

»
3 years ago, # |
  Vote: I like it +8 Vote: I do not like it

Satwik you are working hard but you are wasting a lot of time on easy problems . you must raise the level of questions you are practicing. Try to do atmost 40 questions upto 1200 rating only and spend more time focussing on a bit challanging problems . I hope you will see the improvement very soon . All the best

»
3 years ago, # |
Rev. 3   Vote: I like it +10 Vote: I do not like it

hi, if you don't mind sharing: i am curious to know how you went about solving these problems.

recently i gave up on the approach of solving problems on a progression (like doing a bunch of easy problems -> medium -> harder -> hard, over a couple years or something) because i discovered that i can learn and retain information a lot better after having spent ~20-40 minutes struggling aimlessly on a problem.

when it comes to learning new skills, acetylcholine is a really important neurotransmitter. it aids in the formation of new neurons and neural pathways (learning) and it also helps with long-term and working memory. currently, i am under the impression that you can naturally facilitate the creation of acetylcholine (given that you are under the age ~26) by engaging in novel activities that make you feel confused or frustrated.

here's a summary of my new approach in light of what i just mentioned:

only attempt problems that you don't think you can solve. assuming you don't solve the problem, once you reach the realization that you're probably not going to solve this one: deliberately spend at least 20+ more minutes struggling to solve the problem (this extra time is critical in aiding the learning process that follows). take a break to rest your eyes (10-15 minutes) and then spend however much time is necessary to learn from the top in-contest solutions for the given problem. the next day attempt the problem again.

in the case that you actually did solve the problem the first time around, it is still helpful to check top in-contest solutions.

i recognize that i am only 300 rating at the time of writing this, and that you wanted help from experienced people. here are some resources i have found from experienced people so far:

https://www.codechef.com/wiki/tutorial-dynamic-programming

https://leetcode.com/discuss/general-discussion/1050391/Must-do-Dynamic-programming-Problems-Category-wise

https://drive.google.com/file/d/1J2x8pIYQ3MXANgvzOgBciWd3d79j_Exa/view

https://cses.fi/book/book.pdf

https://cp-algorithms.com/

https://a2oj.com/ namely: https://a2oj.com/Ladders.html

remember that more time does not necessarily lead you to become better at something. what leads you to become better at something is intentional practice.

»
3 years ago, # |
Rev. 4   Vote: I like it -23 Vote: I do not like it

I would like add one point here that quality matters not quantity. For example Blood_Wu_Ren became master after solving 353 problems whereas sgtlaugh became a master after solving 1578 problems and rainboy became master after solving 3250 problems(more than tourist). There is no direct relation between problems and rating. I'm a newbie right now but would like to suggest that you try to figure out where exactly you're making mistakes while solving and as other users suggested try solving problems above your current level.

  • »
    »
    3 years ago, # ^ |
      Vote: I like it +86 Vote: I do not like it

    rainboy's max rating is master because he doesn't want to be grandmaster

    • »
      »
      »
      3 years ago, # ^ |
        Vote: I like it +14 Vote: I do not like it

      I think you forgot to add prefix before grandmaster

    • »
      »
      »
      3 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      orz

      rgnerdplayer is the real orz, he upsolves most of the problems in the contests, even from the time he was expert. True Inspiration!!

      • »
        »
        »
        »
        3 years ago, # ^ |
        Rev. 2   Vote: I like it +3 Vote: I do not like it

        Well, most of the time (especially when I was ~1600) I just read the solution and write something similar.

        • »
          »
          »
          »
          »
          3 years ago, # ^ |
            Vote: I like it 0 Vote: I do not like it

          Exactly what a down-to-earth person would say!!! orz

  • »
    »
    3 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    aren't these mentions unnecessary?

    • »
      »
      »
      3 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      Oh my bad didn't know that it mentioned the user, Fixed it.

      Is there some other way to mention user that doesn't sends them notification?

      • »
        »
        »
        »
        3 years ago, # ^ |
          Vote: I like it +4 Vote: I do not like it

        That is the exact reason why mentions exists. If you just want to link then just link their profile page. like, FoxxKnight

»
3 years ago, # |
  Vote: I like it 0 Vote: I do not like it

Have a try to this CSES

Solutons

And Practice daily (your+200) rating random 3 problems under one hour timing by creating mashup contest in GYM ,

You can use this to select problems but don't see tag while selecting and creating mashup contest at gym.

Best of luck : )

»
3 years ago, # |
  Vote: I like it 0 Vote: I do not like it

In my case, I just followed E869120's guide on how to practice competitive programming and well I gotta say it worked out well, so perhaps you could read through his guide.

»
3 years ago, # |
  Vote: I like it +16 Vote: I do not like it

Relatable.

»
3 years ago, # |
  Vote: I like it 0 Vote: I do not like it

Looking from your submissions, I see a lot of A's and B's. These A's and B's will get you nowhere, that is the harsh truth. You must make it a goal to upsolve till atleast D in a div2 round and the whole Div3 problemset, only then will you significantly improve. Of course everyone is different, as for me I donot practice a lot of stuff from problemset, I only upsolve contest problems as is quite evident from my submission graph (~80 problems lol). Quality matters not quantity.

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

    i don't understand the editorial while solving div2 C,D. what can i do? my rating graph is also very poor. can you suggest me what should i do now?

    • »
      »
      »
      3 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      I suggest you to first improve your theoretical knowledge by reading a book. For example, Competitive Programmer's Handbook is a good one.

    • »
      »
      »
      3 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      You are stuggling with A's and B's, and also looking at your graph, you are quite experienced with CP. I suggest you virtually solve Atcoder beginner contests, they will improve your math skill and implementation skill.

      • »
        »
        »
        »
        3 years ago, # ^ |
          Vote: I like it 0 Vote: I do not like it

        Struggling with A's and B's is one thing. But failing to understand the editorials isn't something that can be improved with just more practice. More likely than not, this indicates gaps in sabbirh654's theoretical knowledge. An editorial explains the idea in just a few sentences, while a book provides pages of more detailed explanations for commonly used algorithms with nice examples.

        • »
          »
          »
          »
          »
          3 years ago, # ^ |
            Vote: I like it 0 Vote: I do not like it

          Books donot teach CP, books merely layout the standard algorithms, CP can only be learnt through lots of problem solving and headbutting with the editorials and problems no matter how hard it seems. I doubt if there is even a single person on this site who would prefer reading books on CP to solving problems.

          • »
            »
            »
            »
            »
            »
            3 years ago, # ^ |
              Vote: I like it 0 Vote: I do not like it

            I doubt if there is even a single person on this site who would prefer reading books on CP to solving problems.

            Okay, let's suppose that somebody doesn't know the theory and has troubles reading editorials (simply because they use all that unfamiliar terminology or refer to unknown things and algorithms). Your recipe is basically to go and try solving even more problems, encounter difficulties again and fail to understand the editorials again.

            There's never a forced choice between reading books (or blogs, comments or whatever else) and solving problems. It's possible to do both.

            • »
              »
              »
              »
              »
              »
              »
              3 years ago, # ^ |
                Vote: I like it 0 Vote: I do not like it

              If you dont understand the editorials, you are simply not trying hard enough.

              • »
                »
                »
                »
                »
                »
                »
                »
                3 years ago, # ^ |
                  Vote: I like it 0 Vote: I do not like it

                What are you talking about? I do understand the editorials. Precisely because I'm reading books and blogs, which explain the relevant theory. While you are saying that you doubt that anyone on this site is doing that.

                Once again. It's sabbirh654, who doesn't understand div2 C and D editorials. Your recommendation is: "go and solve more C and D problems". And my recommendation is: "it's maybe a good idea to read some book about the algorithms typically used in problems C and D, then check the editorials again".

»
3 years ago, # |
  Vote: I like it +1 Vote: I do not like it

Try to solve different types of problems..just don't stuck in one kind of problem.. :)

»
3 years ago, # |
  Vote: I like it 0 Vote: I do not like it

You may try working on your intuition. The problems around 800-1200 doesn't need a lot of algo or implemetation skill, you just need to observe them and try to figure something out. If you can improve the way you think about a problem, i think it'll be helpful.