1: Cauchy Distribution
Probability Density Function
The Cauchy distribution has the probability density function
-
= {1 \ over \ PI} \ left [{\ gamma \ over (X-x_0) ^ 2 + \ gamma ^ 2} \ right], ">
WhereX0 is the location parameter, specifying the location of the peak of the distribution, andGammaIs the scale parameter which specifies the half-width at half-maximum (hwhm ).GammaIs also equal to half the interquartile range and is sometimes called the probable error. cauchy himself exploited such a density function in 1827, with infinitesimal scale parameter, in defining a Dirac delta function (see there ).
Probability Density Function
The purple curve is the standard Cauchy Distribution
The special case whenX0 = 0 andGamma= 1 is calledStandard Cauchy DistributionWith the probability density function
Cumulative Distribution Function
The cumulative distribution function (CDF) is:
Cumulative Distribution Function
2: p-stable distributions
According to the above principle, it is easy to prove that the standard normal distribution is 2-stable.
++ ++ ++
Problem:
1: how to calculate the K value in advance
Obtain a small number of points from dataset at random, calculate one side based on the algorithm, and increase the K value to minimize the computing time by finding a K value.
2: how to store the bucket
Each vertex has a K-element vector. In fact, each element in the vector is of the same nature, but it uses different hash functions. The specific distribution depends on the H1 function.
3: how to ensure accuracy
The manual Manual provides a detailed description of why the author chooses standard normal distribution because the standard normal distribution is 2-stable, which guarantees the accuracy of mathematics.