Editorial for ICPC NWERC 2014 F - Finding Lines

Remember to use this editorial only when stuck, and not to copy-paste code from it. Please be respectful to the problem author and editorialist.
Submitting an official solution before solving the problem yourself is a bannable offence.

Given $n$ ~n~ points in $\mathbb R^2$ ~\mathbb R^2~, is there a straight line passing through at least a $p$ ~p~ fraction of them?

If line exists it is uniquely determined by any two of its points.
Probability that a random point lies on the line is $\ge p$ ~\ge p~.
Repeat $250$ ~250~ times:
1. Pick two distinct points uniformly at random
2. Check if line defined by the points has enough points.
$Pr[\text{false negative}] \approx (1-p^2)^{250} \le (1-\frac{1}{25})^{250} < 4 \times 10^{-5}$ (cheated slightly – it's a bit worse)

Many other ways e.g. divide and conquer

Comments

There are no comments at the moment.