On the number of modes of Gaussian kernel density estimators

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Geshkovski, Borjan, Rigollet, Philippe, Sun, Yihang
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911254079078400
author Geshkovski, Borjan
Rigollet, Philippe
Sun, Yihang
author_facet Geshkovski, Borjan
Rigollet, Philippe
Sun, Yihang
contents We consider the Gaussian kernel density estimator with bandwidth $β^{-\frac12}$ of $n$ iid Gaussian samples. Using the Kac-Rice formula and an Edgeworth expansion, we prove that the expected number of modes on the real line scales as $Θ(\sqrt{β\logβ})$ as $β,n\to\infty$ provided $n^c\lesssim β\lesssim n^{2-c}$ for some constant $c>0$. An impetus behind this investigation is to determine the number of clusters to which Transformers are drawn in a metastable state.
format Preprint
id arxiv_https___arxiv_org_abs_2412_09080
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle On the number of modes of Gaussian kernel density estimators
Geshkovski, Borjan
Rigollet, Philippe
Sun, Yihang
Statistics Theory
Machine Learning
We consider the Gaussian kernel density estimator with bandwidth $β^{-\frac12}$ of $n$ iid Gaussian samples. Using the Kac-Rice formula and an Edgeworth expansion, we prove that the expected number of modes on the real line scales as $Θ(\sqrt{β\logβ})$ as $β,n\to\infty$ provided $n^c\lesssim β\lesssim n^{2-c}$ for some constant $c>0$. An impetus behind this investigation is to determine the number of clusters to which Transformers are drawn in a metastable state.
title On the number of modes of Gaussian kernel density estimators
topic Statistics Theory
Machine Learning
url https://arxiv.org/abs/2412.09080