Selective inference for latent block models

DOI10.1214/21-EJS1853zbMATH Open1471.62265arXiv2005.13273OpenAlexW3167177323MaRDI QIDQ2044409FDOQ2044409

Authors: Taiji Suzuki, Chihiro Watanabe 

Publication date: 9 August 2021

Published in: Electronic Journal of Statistics (Search for Journal in Brave)

Abstract: Model selection in latent block models has been a challenging but important task in the field of statistics. Specifically, a major challenge is encountered when constructing a test on a block structure obtained by applying a specific clustering algorithm to a finite size matrix. In this case, it becomes crucial to consider the selective bias in the block structure, that is, the block structure is selected from all the possible cluster memberships based on some criterion by the clustering algorithm. To cope with this problem, this study provides a selective inference method for latent block models. Specifically, we construct a statistical test on a set of row and column cluster memberships of a latent block model, which is given by a squared residue minimization algorithm. The proposed test, by its nature, includes and thus can also be used as the test on the set of row and column cluster numbers. We also propose an approximated version of the test based on simulated annealing to avoid combinatorial explosion in searching the optimal block structure. The results show that the proposed exact and approximated tests work effectively, compared to the naive test that did not take the selective bias into account.

Full work available at URL: https://arxiv.org/abs/2005.13273

Recommendations

• Goodness-of-fit test for latent block models
• Model selection for Gaussian latent block clustering with the integrated classification likelihood
• Estimation and selection for the latent block model on categorical data
• Block clustering with collapsed latent block models
• Convergence of the groups posterior distribution in latent or stochastic block models

zbMATH Keywords

latent block modelrelational data analysisselective inferenc

Mathematics Subject Classification ID

Parametric hypothesis testing (62F03) Classification and discrimination; cluster analysis (statistical aspects) (62H30)

Cites Work

• Latent block model for contingency table
• Model selection for Gaussian latent block clustering with the integrated classification likelihood
• Bsig: evaluating the statistical significance of biclustering solutions
• Optimization by simulated annealing
• Block clustering with Bernoulli mixture models: comparison of different approaches
• Title not available (Why is that?)
• Block clustering with collapsed latent block models
• Exact post-selection inference, with application to the Lasso
• Valid post-selection inference
• Hypothesis Testing for Automated Community Detection in Networks
• A goodness-of-fit test for stochastic block models
• Selective inference with a randomized response
• Mathematical Statistics
• Poisson random fields for dynamic feature models
• Cooling Schedules for Optimal Annealing
• Thermodynamical approach to the travelling salesman problem: An efficient simulation algorithm
• Uniform asymptotic inference and the bootstrap after model selection
• Biclustering via sparse singular value decomposition
• Goodness-of-fit test for latent block models
• Convex biclustering

Cited In (4)

• Selective inference for clustering with unknown variance
• Goodness-of-fit test for latent block models
• Likelihood-based model selection for stochastic block models
• Estimation and selection for the latent block model on categorical data

Uses Software

• Bsig
• MovieLens

This page was built for publication: Selective inference for latent block models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2044409)