ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 1804.02756
61
8
v1v2v3v4 (latest)

An adaptive multiclass nearest neighbor classifier

8 April 2018
Nikita Puchkin
V. Spokoiny
ArXiv (abs)PDFHTML
Abstract

We consider a problem of multiclass classification, where the training sample Sn={(Xi,Yi)}i=1nS_n = \{(X_i, Y_i)\}_{i=1}^nSn​={(Xi​,Yi​)}i=1n​ is generated from the model P(Y=m∣X=x)=ηm(x)\mathbb P(Y = m | X = x) = \eta_m(x)P(Y=m∣X=x)=ηm​(x), 1≤m≤M1 \leq m \leq M1≤m≤M, and η1(x),…,ηM(x)\eta_1(x), \dots, \eta_M(x)η1​(x),…,ηM​(x) are unknown α\alphaα-Holder continuous functions.Given a test point XXX, our goal is to predict its label. A widely used k\mathsf kk-nearest-neighbors classifier constructs estimates of η1(X),…,ηM(X)\eta_1(X), \dots, \eta_M(X)η1​(X),…,ηM​(X) and uses a plug-in rule for the prediction. However, it requires a proper choice of the smoothing parameter k\mathsf kk, which may become tricky in some situations. In our solution, we fix several integers n1,…,nKn_1, \dots, n_Kn1​,…,nK​, compute corresponding nkn_knk​-nearest-neighbor estimates for each mmm and each nkn_knk​ and apply an aggregation procedure. We study an algorithm, which constructs a convex combination of these estimates such that the aggregated estimate behaves approximately as well as an oracle choice. We also provide a non-asymptotic analysis of the procedure, prove its adaptation to the unknown smoothness parameter α\alphaα and to the margin and establish rates of convergence under mild assumptions.

View on arXiv
Comments on this paper