Alex7's blog

Training?
By Alex7, 9 years ago, In English

Would it be practical to make a suggested list of problems for training based on rating ranges?

  • Vote: I like it
  • +29
  • Vote: I do not like it

»
9 years ago, # |
  Vote: I like it +39 Vote: I do not like it

Well, since you're almost red maybe you could compile a list for us lower-rated peasants?

→ Reply
»
9 years ago, # |
← Rev. 2   Vote: I like it +8 Vote: I do not like it

The question can be answered using "YES" or "NO" but I don't have a brute-force solution (try training based on rating ranges on a lot of people and try training based on other factors) to state but I may use some heuristics.

You're going to join IOI, as far as I know, so you should train on the IOI problems. Unfortunately, CodeForces doesn't include interaction problems and Codeforces is more biased in favor of ICPC. I think it'd be more practical in this period to focus more on IOI-style contests and problems. Codechef lunchtime will be held the day after tomorrow and the problems will be partially graded.

In a general case, I suppose you should learn new topics or solve problems on your weak points. It's not really beneficial if you know which topic, the problem involves, in advance, the problem is already spoiled. You may ask someone to prepare for you an unordered-list of problems on various topics that you find them somehow hard or you need to practice them more. The difficulty rating is subjective, you may -rarely- find Div1-D easier than Div1-C or even Div1-B and you may be familiar with the harder problem (for example: I was familiar with Div1-C in round 305 but not with Div1-B), so it's more reasonable to train on the topics you didn't master than the estimated level of problems you may haven't -or may have- reached, yet.

→ Reply
»
9 years ago, # |
  Vote: I like it +23 Vote: I do not like it

Just because two person lie on same rating range, doesn't necessarily mean that they have equivalent set of knowledge. So considering that, it would be difficult to make suggestions right? Perhaps we could use something else to sort people.

→ Reply
»
9 years ago, # |
← Rev. 2   Vote: I like it +7 Vote: I do not like it

If you order CF's problemset by solved then you should train on problem number (your_rating — 1500)*2, it should roughly be your limit to what you can comfortably solve. I started with competitive programming half a year ago and I have solved over a thousand problems here on CF since then so I know what they are like.

By the way, there is no use in trying to solve too hard problems, ideally you should train on problems that you can solve slowly or that you would likely get a WA on in a competition. It could be due to some gotcha you missed (most of the later "easy" problems are like this, (extreme example)), you aren't really sure about an algorithm or that you make too many mistakes with your implementations.

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

    so 350'th problem isn't easy to solve?

    → Reply
    • »
      »
      »
      9 years ago, # ^ |
      ← Rev. 2   Vote: I like it 0 Vote: I do not like it

      Ah sorry, should be (your_rating — 1500)*2, not (your_rating — 1500)/2.

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

        sounds kinda reasonable but they are just medium problems that i could most likely solve in a contest

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

          If the risk of failing is above 20% or if it takes more than 15 minutes to solve then I think that it is worth solving because they will otherwise become a huge mental drain during competitions.

          Building up my basics like this is why I could take first place during round 1c in this years CodeJam. It doesn't mean that I'm good since a lot of the best weren't in it and since all the problems were easy, but at least it means that I can do those kind of problems better than most.

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

            well i agree with you

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

    That's not a good formula for green and gray coders :D

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

    1000+ problems in only 6 months.. WOW

    → Reply
  • »
    »
    9 years ago, # ^ |
      Vote: I like it -6 Vote: I do not like it

    Wow, just looked at your submission history, you da real mvp Master Joh, I will follow your ways.

    → Reply