• Corpus ID: 88520170

The functional mean-shift algorithm for mode hunting and clustering in infinite dimensions

  title={The functional mean-shift algorithm for mode hunting and clustering in infinite dimensions},
  author={Mattia Ciollaro and Christopher R. Genovese and Jing Lei and Larry A. Wasserman},
  journal={arXiv: Methodology},
We introduce the functional mean-shift algorithm, an iterative algorithm for estimating the local modes of a surrogate density from functional data. We show that the algorithm can be used for cluster analysis of functional data. We propose a test based on the bootstrap for the significance of the estimated local modes of the surrogate density. We present two applications of our methodology. In the first application, we demonstrate how the functional mean-shift algorithm can be used to perform… 

Nonparametric Clustering of Functional Data Using Pseudo-Densities

We study nonparametric clustering of smooth random curves on the basis of the L2 gradient flow associated to a pseudo-density functional and we show that the clustering is well-defined both at the

A review of mean-shift algorithms for clustering

The theory and practice behind clustering based on kernel density estimates and mean-shift algorithms are described and applications to image segmentation, manifold denoising and multivalued regression are discussed.

Evaluating the complexity of some families of functional data

This paper studies the complexity of a functional data set drawn from particular processes by means of a two-step approach using a new graphical tool based on a nonparametric kNN estimation of small ball probability.

Functional summaries of persistence diagrams

The definition of persistence landscape functions is generalized, several theoretical properties of the persistence functional summaries are established, and their performance in the context of classification using simulated prostate cancer histology data is demonstrated.

An empirical analysis of different sparse penalties for autoencoder in unsupervised feature learning

Experimental study on MNIST, CIFAR-10, SVHN, OPTDIGITS and NORB datasets reveals that all these penalties achieve sparse representation and outperforms representations learned by pure autoencoder on classification performance and sparseness of feature vectors.



Mean Shift, Mode Seeking, and Clustering

  • Yizong Cheng
  • Computer Science
    IEEE Trans. Pattern Anal. Mach. Intell.
  • 1995
Mean shift, a simple interactive procedure that shifts each data point to the average of data points in its neighborhood is generalized and analyzed and makes some k-means like clustering algorithms its special cases.

Nonparametric estimation of the mode of a distribution of random curves

Methods for density and mode estimation when data are in the form of random curves are introduced based on finite dimensional approximations via generalized Fourier expansions on an empirically chosen basis.

Bandwidth Selection for Mean-shift based Unsupervised Learning Techniques: a Unified Approach via Self-coverage

This paper proposes to use a so-called self-coverage measure as a general device for bandwidth selection in this context and shows how a bandwidth h will be favorable if a high proportion of data points falls within circles or ``hypertubes"; of radius h centered at the fitted object.

Data-driven density derivative estimation, with applications to nonparametric clustering and bump hunting

This paper presents the first fully automatic, data-based bandwidth selectors for multivariate kernel density derivative estimators by synthesizing recent advances in matrix analytic theory which allow mathematically and computationally tractable representations of higher order derivatives of multivariate vector valued functions.

The variable bandwidth mean shift and data-driven scale selection

The sample point estimator is defined, prove its convergence, and show its superiority over the fixed bandwidth procedure, and an alternative approach for data-driven scale selection which imposes a local structure on the data is studied.

Mean Shift: A Robust Approach Toward Feature Space Analysis

It is proved the convergence of a recursive mean shift procedure to the nearest stationary point of the underlying density function and, thus, its utility in detecting the modes of the density.

A reliable data-based bandwidth selection method for kernel density estimation

The key to the success of the current procedure is the reintroduction of a non- stochastic term which was previously omitted together with use of the bandwidth to reduce bias in estimation without inflating variance.

On the convergence of the mean shift algorithm in the one-dimensional space

A note on density mode estimation

Persistence-Based Clustering in Riemannian Manifolds

A clustering scheme that combines a mode-seeking phase with a cluster merging phase in the corresponding density map, and whose output clusters have the property that their spatial locations are bound to the ones of the basins of attraction of the peaks of the density.