Cautious active clustering

DOI10.1016/J.ACHA.2021.02.002MaRDI QIDQ2036490zbMATH OpenOpenAlexFDO

Authors Alexander Cloninger, H. N. Mhaskar

Publication date 29 June 2021

Published in Applied and Computational Harmonic Analysis (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2008.01245

zbMATH Keywords

active clusteringhierarchical clusteringhyperspectral pixel classification

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30) Image analysis in multivariate analysis (62H35)

Abstract: We consider the problem of classification of points sampled from an unknown probability measure on a Euclidean space. We study the question of querying the class label at a very small number of judiciously chosen points so as to be able to attach the appropriate class label to every point in the set. Our approach is to consider the unknown probability measure as a convex combination of the conditional probabilities for each class. Our technique involves the use of a highly localized kernel constructed from Hermite polynomials, in order to create a hierarchical estimate of the supports of the constituent probability measures. We do not need to make any assumptions on the nature of any of the probability measures nor know in advance the number of classes involved. We give theoretical guarantees measured by the $F$-score for our classification scheme. Examples include classification in hyper-spectral images and MNIST classification.

Recommendations

• Two faces of active learning
• Clustering of measures via mean measure quantization
• A probabilistic theory of clustering
• scientific article; zbMATH DE number 5968861
• scientific article; zbMATH DE number 1140602

Cites work

• A Fourier-invariant method for locating point-masses and computing their attributes
• A guided tour of Chernoff bounds
• Data-driven atomic decomposition via frequency extraction of intrinsic mode functions
• Diffusion polynomial frames on metric measure spaces
• scientific article; zbMATH DE number 823880 (Why is no real title available?)
• Local approximation using Hermite functions
• Minimax analysis of active learning
• On local smoothness classes of periodic functions
• On the detection of singularities of a periodic function
• On the representation of smooth functions on the sphere using finitely many bits
• Performance guarantees for hierarchical clustering
• Polynomial operators and local smoothness classes on the unit interval. II.
• Representation of functions on big data associated with directed graphs
• Representation of functions on big data: graphs and trees
• Signal decomposition and analysis via extraction of frequencies
• Spatial-spectral operator theoretic methods for hyperspectral image classification
• Super-Resolution Limit of the ESPRIT Algorithm

Cited in

(5)

• Clustering of measures via mean measure quantization
• Applied harmonic analysis and data science. Abstracts from the workshop held November 28 -- December 4, 2021 (hybrid meeting)
• Poisson Reweighted Laplacian Uncertainty Sampling for Graph-Based Active Learning
• An eigenfunction approach to conversion of the Laplace transform of point masses on the real line to the Fourier domain
• A signal separation view of classification

Describes a project that uses

Uses Software

• MNIST

This page was built for publication: Cautious active clustering

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2036490)