Approximate Nearest Centroid Embedding for Kernel k-Means
نویسندگان
چکیده
This paper proposes an efficient embedding method for scaling kernel k-means on cloud infrastructures. The embedding method allows for approximating the computation of the nearest centroid to each data instance and, accordingly, it eliminates the quadratic space and time complexities of the cluster assignment step in the kernel k-means algorithm. We show that the proposed embedding method is effective under memory and computing power constraints, and that it achieves better clustering performance compared to other approximations of the kernel kmeans algorithm.
منابع مشابه
Scalable Kernel k-Means via Centroid Approximation
Although kernel k-means is central for clustering complex data such as images and texts by implicit feature space embedding, its practicality is limited by the quadratic computational complexity. In this paper, we present a novel technique based on scalable centroid approximation that accelerates kernel k-means down to a sub-quadratic complexity. We prove near-optimality of our algorithm, and e...
متن کاملAsymptotic Behaviors of Nearest Neighbor Kernel Density Estimator in Left-truncated Data
Kernel density estimators are the basic tools for density estimation in non-parametric statistics. The k-nearest neighbor kernel estimators represent a special form of kernel density estimators, in which the bandwidth is varied depending on the location of the sample points. In this paper, we initially introduce the k-nearest neighbor kernel density estimator in the random left-truncatio...
متن کاملPartitioning Well-Clustered Graphs: Spectral Clustering Works!
In this work we study the widely used spectral clustering algorithms, i.e. partition a graph into k clusters via (1) embedding the vertices of a graph into a low-dimensional space using the bottom eigenvectors of the Laplacian matrix, and (2) partitioning the embedded points via k-means algorithms. We show that, for a wide class of graphs, spectral clustering algorithms give a good approximatio...
متن کاملSpectral methods for image clustering
The clustering problem is to assign labels to points in order to group them in a structurally meaningful way. This is often accomplished by defining cluster centroids in the vector space and assigning points to the cluster with the nearest centroid, as the k-means algorithm does. But this approach does not work for clusters that have unusual shapes, particularly if the clusters are interwoven o...
متن کاملFast k-means based on KNN Graph
In the era of big data, k-means clustering has been widely adopted as a basic processing tool in various contexts. However, its computational cost could be prohibitively high as the data size and the cluster number are large. It is well known that the processing bottleneck of k-means lies in the operation of seeking closest centroid in each iteration. In this paper, a novel solution towards the...
متن کامل