Блог пользователя bayleef

Автор bayleef, история, 9 лет назад, По-английски

<Copy-paste from shashank21j>

Hi!

Ad Infinitum is back with 14th Edition, a contest restricted to mathematics domain, held on HackerRank.

Register at https://www.hackerrank.com/infinitum14
The contest commences on 15th January 15:30 UTC. You are allowed to enter the contest anytime, for tie breaking the timer will start when you view the first challenge, which allows you to start late at your convenience, but once started try to finish as fast as possible to be on top of the leaderboard :)

It's a 2 days contest with 8 problems (3 Easy(Tutorial), 3 Medium, 2 Hard)
Scoring Distribution: 20 25 30 40 60 80 100 140

Top 10 on leaderboard gets Cool HackerRank T shirt,

Detailed editorials will be available by the end of contest :) I suggest you to try all challenges and at the end of contest understand the solutions.

GL&HF

</Copy-paste from shashank21j>

Hope you find problems interesting, especially the last one :)

  • Проголосовать: нравится
  • +70
  • Проголосовать: не нравится

»
9 лет назад, # |
  Проголосовать: нравится +28 Проголосовать: не нравится

I'm one of the setters.

It seems the problem set is unusually hard (and interesting!) this time around. Should we extend the contest by 24 hours?

»
9 лет назад, # |
  Проголосовать: нравится +13 Проголосовать: не нравится

In editorial of problem D O(n) solution is described. However there exists a solution in compelxity O(some very small function(n, ε)) http://ideone.com/XJirM2 (probably (where ε is demanded precision). Just note that from some point E and V grows almost linearly, because probabilities of getting i in j-th roll tend really fast to stationary distribution.

»
9 лет назад, # |
  Проголосовать: нравится +13 Проголосовать: не нравится

Another way to solve "Dice Stats" is to use the law of total variance.

»
9 лет назад, # |
  Проголосовать: нравится +5 Проголосовать: не нравится

I really want to participate the next one of Ad Infinitum. When is that?

»
9 лет назад, # |
  Проголосовать: нравится +5 Проголосовать: не нравится

It looks like in problem Random Number Generator you didn't strictly prove that it's optimal to set some probabilities to zero.