Clustering is a widely used technique in machine learning, however, relatively little research in consistency of clustering algorithms has been done so far. In this paper we investigate the consistency of the regularized spectral clustering algorithm, which has been proposed recently. It provides a ...
Consistency of Cheeger and ratio graph cuts,preprint 分析了谱聚类的consistency A variational approach to the consistency of spectralclustering,Applied and Computational Harmonic Analysis,preprint,published 以及他们最近推广到了lp和error rate了[我还没看 Error estimates for spectral convergence of the graph L...
Existing works tend to execute the subspace learning and spectral clustering in two separated steps, without consideration of the fact that these two steps highly depend on each other. Definition: Exclusivity: Exclusivity between two matrix ∈ ℝn×n and ∈ ℝn×n is defined as ...
Single cell RNA sequencing is a promising technique to determine the states of individual cells and classify novel cell subtypes. In current sequence data analysis, however, genes with low expressions are omitted, which leads to inaccurate gene counts an
the representation learning and spectral clustering in two separated steps, which also limits its performance. Based on these observations, we introduce a novel position-aware exclusivity term to exploit the complementary information and an indicator consistency term to unify the processing of s...
Another interesting fact is that the formula of the LC-Stat is similar to the objective function in spectral clustering [23], which aims at finding a best cut for a given graph structure. Spectral clustering does not make strong assumptions on the statistics of the data and shows good ...
graph-based clustering techniques, spectral clustering algorithms (SC) such as NJW [27] firstly construct a similarity matrix based on the original data and then utilize the eigenvectors of the normalized Laplacian matrix derived from the constructed similarity matrix to perform data clustering. More...
In this paper, we give a broad overview of the intersection of partial differential equations (PDEs) and graph-based semi-supervised learning. The overview is focused on a large body of recent work on PDE continuum limits of graph-based learning, which have been used to prove well-posedness ...
Our work is inspired by the spectral characteristics of the low-light images and primarily utilizes the spectral perturbations to establish the training constraints. Specifically, we first investigate the invariant and equivariant components for low-light enhancement under spectral perturbations. Based on ...
This consistency could be expressed either by maintaining the constancy of spectral properties and/or the constancy in the produced duration of segments. The current study sought evidence for use of the above two strategies in Lombard speech production (i.e., more distributed vowels and/or more ...