F Nous contacter


0

Documents  62G07 | enregistrements trouvés : 3

O
     

-A +A

Sélection courante (0) : Tout sélectionner / Tout déselectionner

P Q

We present two related contributions of independent interest: high-probability finite sample rates for $k$-NN density estimation, and practical mode estimators ­ based on $k$-NN ­ which attain minimax-optimal rates under surprisingly general distributional conditions.

$k$-nearest neighbor ($k$-NN) - $k$-NN density rates - mode estimation

62G07

Consider a sample of points drawn from some unknown density on $R^d$. Assume the only information we have about the sample are the $k$-nearest neighbor relationships: we know who is among the $k$-nearest neighors of whom, but we do not know any distances between points, nor the point coordinates themselves. We prove that as the sample size goes to infinty, it is possible to reconstruct the underlying density p and the distances of the points (up to a multiplicative constant).

$k$-nearest neighbor graph - random geometric graph - ordinal embedding
Consider a sample of points drawn from some unknown density on $R^d$. Assume the only information we have about the sample are the $k$-nearest neighbor relationships: we know who is among the $k$-nearest neighors of whom, but we do not know any distances between points, nor the point coordinates themselves. We prove that as the sample size goes to infinty, it is possible to reconstruct the underlying density p and the distances of the points (up ...

62G07 ; 62G30 ; 68R10

Nuage de mots clefs ici

Z