Hierarchical latent class models for cluster analysis

By Nevin Lianwen Zhang. Journal of machine learning research, 2002

302 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/1b3c3c7064d2c628e78095e77685d216caa90d86

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Sum-product networks: A new deep architecture (2011, background): “In this sense, sum-product networks are more general than both hierarchical mixture models [32] and thin junction trees [5].”
Greedy Learning of Binary Latent Trees (2011, methodology): “However, as is well known in phylogenetics (see, e.g., [6]), the root can be “walked” around the tree without changing the joint distribution; there is an equivalence class of directed trees which corresponds to one undirected tree (see also discussion in [ 21, Section 3.2 ]).”
Kernel Embeddings of Latent Tree Graphical Models (2011, methodology): “[27] proposed a search heuristic for hierarchical latent class models by defining a series of local search operations and using EM to compute the likelihood of candidate structures.”
Bayesian Networks and Influence Diagrams: A Guide to Construction and Analysis (2007, background): “1997) appears as a natural extension of the naive Bayes model, while the hierarchical naive Bayes model (Zhang 2004) is another extension of the naive Bayes model.”
Learning the Structure of Linear Latent Variable Models (2006, background): “Algorithms such as the ones by Elidan et al. (2000) and Zhang (2004) are designed to effectively perform density estimation, which is a very different problem, even if good density estimators provide one possible causal model compatible with the data.”
Differential Privacy for Deep and Federated Learning: A Survey (2022, methodology): “The approach is based on the latent tree model (LTM) [100] and contains three main steps: 1) generating latent attributes by condensing original attributes and adding Laplace noise to guarantee (cid:15) -DP, 2) quantifying the correlation (i.e., mutual information) between any two latent…”