There exist a wide variety of distributions over trees with infinitely many nodes, including the nested Chinese restaurant process (Blei et al., 2004), the Dirichlet diffusion tree (Neal, 2003), and Kingman’s coalescent (Kingman, 1982). These models differ from the TSSBP in that data can only be associated with a leaf node, or equivalently a full path from root to leaf. We chose to base our clu...

The analysis of the submm anisotropies that will be mapped by the forthcoming map and planck satellites requires careful foreground subtraction before measuring CMB fluctuations. Among these, the foreground due to IR/submm thermal radiation from dusty sources was poorly known until recent observational breakthroughs began unveiling the properties of these objects. We hereafter briefly review th...

In this paper we perform an hierarchical clustering in high – dimensional spaces, without first applying any space reduction. Instead, in each step of the algorithm we perform a soft feature selection, witch does not have to be shared among all input elements. The main goal is to correctly identify the patterns that underly in the data. The proposed algorithm is applied, with promising results,...

This paper describes a hierarchical clustering of musical signals based on information derived from spectral and bispectral acoustic distortion measures. This clustering reveals the ultra metric structure that exists in the set of sounds, with a clear interpretation of the distances between the sounds as the statistical divergence between the sound models. Spectral, bispectral and combined clus...

This article presents a probabilistic hierarchical clustering model for morphological segmentation. In contrast to existing approaches to morphology learning, our method allows learning hierarchical organization of word morphology as a collection of tree structured paradigms. The model is fully unsupervised and based on the hierarchical Dirichlet process (HDP). Tree hierarchies are learned alon...

Hamilton et al. have suggested an invaluable scaling formula which describes how the power spectra of density fluctuations evolve into the nonlinear regime of hierarchical clustering. This paper presents an extension of their method to low-density universes and universes with nonzero cosmological constant. We pay particular attention to models with large negative spectral indices, and give a sp...

Last time, we introduced the task of hierarchical clustering, in which we aim to produce nested clusterings that reflect the similarity between clusters. This contrasts sharply with our former discussion of “flat” or structureless clustering methods like k-means which do not model relationships between clusters. In this lecture, we will continue our discussion of the standard model-free approac...

Current methods for hierarchical clustering of data either operate on features of the data or make limiting model assumptions. We present the hierarchy discovery algorithm (HDA), a model-based hierarchical clustering method based on explicit comparison of joint distributions via Bayesian network learning for predefined groups of data. HDA works on both continuous and discrete data and offers a ...

In view of the extensive evidence of tight inter-relationships between spheroidal galaxies (and galactic bulges) with massive black holes hosted at their centers, a consistent model must deal jointly with the evolution of the two components. We describe one such model, which successfully accounts for the local luminosity function of spheroidal galaxies, for their photometric and chemical proper...

This paper presents a new approach to agglomerative hierarchical clustering. Classical hierarchical clustering algorithms are based on metrics which only consider the absolute distance between two clusters, merging the pair of clusters with highest absolute similarity. We propose a relative dissimilarity measure, which considers not only the distance between a pair of clusters, but also how dis...