algorithm - NP-Hard? Algorithmic complexity of online poker collusion detection? -


What is the best way to describe the algorithmic complexity of collusion identity for the ten million-player online poker site?

Suppose (I do not think this perception makes a difference, so feel free to ignore them, but just to make it clear):

  • Site There are over 10,000,000 registered users.
  • These players have played 5 billion hands in total.
  • This information is only "Master Hand History Database" for the information given to you site, in which all player holes are included in the card and betting work for each hand.
  • In other words, you can not take shortcuts like checking the IP address, looking for unusual rake / profit pattern, and so forth.
  • Assume that you have been given a function which, when actually N (where N is between 2 and 10) is passed to a group of players, TRUE returns if all of the group Players have worked together if anything but all players are not colloidals, then FALSE gives the function. A return value of TRUE is done with 75% trust (for example).

Your job is to prepare a complete list of those players, which is in line with the complete list of players, I am recently associated with this problem as NP-Hard Have heard, but is this right? Sometimes we call things "NP" or "NP-Hard" which are simply "tough".

Thank you!

I immediately see that there is force-force approach:

  Set Colugues = New Set (); (Player P1: All Players) {Player Player 2: AllPlayer} {if (! P1Ajel (P2) and Heldlded (P1, P2)) {colluders.add (p1); Colluders.add (P2); }}}  

I do not see any point to call, with logic greater than 2, because it can give wrong negative. I think that it depends on how expensive this work is, but the above results in O (n ^ 2) are called to call (N-player number). The function seems to be O (m), where I have the number of games that they played together. Thus, under the algorithm O (N ^ 3), looks well . To be NP-hard, you can "l have a problem h np-hard if and if only one np-full problem is l, then polynomial time can be turing-reducing to h [...] in other words With an oracle for H, an oracle machine should be resolved in polynomial time. " I have studied NP-complete problems (for example 3-SAT, Traveling salesman problem etc.) and I do not know how you will prove it. But then, it looks like suspicion.


Comments

Popular posts from this blog

python - Overriding the save method in Django ModelForm -

html - CSS autoheight, but fit content to height of div -

qt - How to prevent QAudioInput from automatically boosting the master volume to 100%? -