gm_monkey's blog

By gm_monkey, history, 11 hours ago, In English

Hi! I’m looking for some feedback from people earlier in their CP journey than me :)

I’ve been doing CP for a few years, mostly olympiads and CF. When I got stuck, it was usually on dumb things, like debugging, giving up on an idea too early, or not knowing what to practice next. Nobody told me that was the problem, I just figured it out slowly.

I’m curious what it is for you. What’s the thing slowing you down most right now? And if someone sat with you for an hour, what would you want them to do?

Higher rated people, what do you wish you’d known earlier?

Any answer helps, even one line. Thanks!

  • Vote: I like it
  • +4
  • Vote: I do not like it

»
11 hours ago, hide # |
 
Vote: I like it +4 Vote: I do not like it

I would ask him to teach me how to debug a solution which gets wrong answer on test 2, when we feel the code and the ideas are correct. It always happens that I almost got the right ideas and missed a silly thing.

I've also realised that even after getting the key ideas, I tend to overcomplicate the coding part, and miss small implementation tricks (for ex, iterating from the back of array instead of front or sorting in reverse order )

  • »
    »
    11 hours ago, hide # ^ |
     
    Vote: I like it 0 Vote: I do not like it

    For wa 2, i tend to test with random tests manually. Finding the test that was wrong is the mostly hard part for me. AND PLEASE, don’t look at the wrong test. In no olympiad or contest, you wont see it. Maybe if you are desperate.

    If manual random testing doesn’t work, I create a simple random generator and make a obviously correct brute force. And compare the results. This works like 90% of the time. And you test the brute force by just sending it. Bc most of the time getting time limit in order to get ac is worth it. In OI problems, you even get points for the subtasks.

    After finding the test that was giving wrong answer, you just need to cout every single thing that you think can be the bug.

    But not finding and not being able to fix is also a normal thing. That is why we learn from them. I hope this was an answer good enough to help.

»
10 hours ago, hide # |
 
Vote: I like it +1 Vote: I do not like it

Implementation. I'm so freaking lazy, I just make a solution sketch most of the time. I think once my implementations stops being terrible I'll perchance make it to the big leagues instead of wa1 hell

»
8 hours ago, hide # |
 
Vote: I like it 0 Vote: I do not like it

Being able to solve some high-rated problems but can't solve problems near my rating stably.

»
8 hours ago, hide # |
 
Vote: I like it 0 Vote: I do not like it

i would say there are some times you don't know if you are improving or not cuz you can't solve more problems in contest or even you solve less that you did before , and that will make you question everything you are doing and trying to change everything you do constantly

  • »
    »
    7 hours ago, hide # ^ |
     
    Vote: I like it 0 Vote: I do not like it

    I think as long as you do something that makes your brain work, you’ll improve at something, even if it’s just reading an editorial and really understanding it. And judging your progress by one contest isn’t always the best idea, because your performance depends heavily on luck: your mood, the topics of the problems, and so much more. Your result is a product of you and your environment, and you can’t fully control your environment. No one can. But you still have to try. You can’t win if you don’t fight, so fight to the end.

  • »
    »
    7 hours ago, hide # ^ |
     
    Vote: I like it 0 Vote: I do not like it

    I can see your rating is slowly increasing but figure out what brings you so many WAs and fix that

»
7 hours ago, hide # |
 
Vote: I like it 0 Vote: I do not like it

The toughest—and, frankly, the most "nasty"—topic is dynamic programming. In slightly more advanced problems, the underlying ideas are far from obvious; as far as I can tell, the only way to master them is to just put in the hours—lots of them. Graphs, on the other hand, are the easiest: while there are many variations and a lot of theory, even problems with a 2000 rating can be quite intuitive. Proving the correctness of a solution is also difficult, especially when the problem seems simple and self-evident. But it’s even worse when you have absolutely no idea how to even approach a problem. Technology has been a huge help for me—I use ChatGPT as a sort of personal tutor or mentor. It’s easy to find information on unfamiliar topics, and if you get stuck, the AI ​​can step in. Best of all, it can curate a collection of problems on any topic, arranged by increasing difficulty! My biggest challenge is anxiety. With Olympiads and the Unified State Exam (EGE) on the horizon—and given that I’m also quite lazy—it’s tough. I haven’t managed to pass any of the selection rounds I’ve attempted so far, though I’m hoping that’s just for now)

»
7 hours ago, hide # |
 
Vote: I like it 0 Vote: I do not like it

IMPLEMENTATION...

»
6 hours ago, hide # |
 
Vote: I like it 0 Vote: I do not like it

For me, construction problems are toughest. The editorials for those problems are always like: "here's a magical construction that works because it works." They don't seem approachable by logic.

I occasionally fail to solve easy construction problems, and I feel like I'm an idiot every time I do so...

»
93 minutes ago, hide # |
 
Vote: I like it 0 Vote: I do not like it

Probably thinking, and learning how to think properly. For lower rated problems, usually if you have the right approach and covered all edge cases while thinking, implementation is straightforward. Knowing when to stop pursuing dead ideas, proving solutions quickly, turning the current problem into a past problem, reducing a problem into a simpler problem, coming up with good ideas at all.

»
18 minutes ago, hide # |
 
Vote: I like it 0 Vote: I do not like it

Solving problems is pretty annoying