anup.kalbalia's blog

By anup.kalbalia, history, 9 years ago, In English

Hello Codeforces Community!

CodeChef invites you to participate in CodeChef August Long Challenge at http://www.codechef.com/AUG15. As always, the long contests will start on first Friday of every month and last for 10 days. This ensures that two weekends are covered.

Time: 7th August 2015 (1500 hrs) to 17th August 2015 (1500 hrs). (Indian Standard Time — +5:30 GMT) — Check your timezone.

Details: https://www.codechef.com/AUG15

Registration: Just need to have a CodeChef handle to participate. For all those, who are interested, are requested to register in order to participate.

Prizes:

For Indians:

1st prize: INR 12000

2nd prize: INR 8000

For Rest of the world:

1st prize: $400

2nd prize: $300

In addition to the above prizes. Top 10 global, Top 20 Indian, Top 5 school (Indian), Top 5 school (Global), Top 3 challenge problem solvers Indian and Global (apart from the winners) get cool CodeChef goodies. Starting with August Challenge 2015, top 5 girl coders will be provided with goodies, given that they are among top 100 in the final ranklist (this will be irrespective of the region).

It promises to deliver on an interesting set of algorithmic problems with something for all.

Good Luck! Hope to see you participating!!

  • Vote: I like it
  • +8
  • Vote: I do not like it

| Write comment?
»
9 years ago, # |
  Vote: I like it +8 Vote: I do not like it

I think you forgot to change the first paragraph from July to August.

»
9 years ago, # |
Rev. 2   Vote: I like it +20 Vote: I do not like it

I wanted to post my thoughts for problem SCLUSTER under its editorial, but your CAPTCHA-script looks like an iframe loading from HTTP-page, while the whole site is in HTTPS, so my browser blocks loading with an error

Mixed Content: The page at 'https://discuss.codechef.com/questions/73728/scluster-editorial#fmanswer' was loaded over HTTPS, but requested an insecure script 'http://api.recaptcha.net/challenge?k=6Ld28wYTAAAAAIAqh2vqc4dsCE3bz-oDorke5j4E'. This request has been blocked; the content must be served over HTTPS.

and I'm not able to post anything :( Both Firefox 32.0 and Chrome 41.0.2272.118 for Linux Mint 17.0 show such behavior. Please fix that.

So, I'll post my thoughts here.

My algorithm consists mostly from two things:

1) building a connected component affecting as many low-li people as possible.

2) rearranging the rest of the people in this connected component in order to minimize A by using Hungarian algorithm

After that I experimented with procedures that minimize value of B, but I didn't succeed in it. I think, the metric 1000 * A + 10 * B is bad because even if a person has large value of li = 25, then moving it by 6 cells will cost 1000 * 6 / 25 = 240 from 1000A, and possible profit of its rearrangement is no higher than |25 - 1| * 10 = 240 from 10B. In fact profit is much lower because there is a denominator that equals to a degree of destination cell, so it's not really an effective optimization.