Likelihood estimation of sparse topic distributions in topic models and its applications to Wasserstein document distance calculations

DOI10.1214/22-AOS2229MaRDI QIDQ2112814

Seth Strimas-Mackey, Xin Bing, Florentina Bunea, Marten H. Wegkamp

Publication date: 12 January 2023

Published in: The Annals of Statistics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2107.05766

zbMATH Keywords

maximum likelihood estimationadaptive estimationmultinomial distributionmixture modelminimax estimationnonnegative matrix factorizationsparse estimationhigh-dimensional estimationtopic modelsanchor words

Mathematics Subject Classification ID

Estimation in multivariate analysis (62H12) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Point estimation (62F10)

Related Items

Assigning topics to documents by successive projections

Uses Software

• word2vec
• Sentence-BERT

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• A fast algorithm with minimax optimal guarantees for topic models with an unknown number of topics
• Empirical optimal transport on countable metric spaces: distributional limits and statistical applications
• Sharp asymptotic and finite-sample rates of convergence of empirical measures in Wasserstein distance
• Simultaneous analysis of Lasso and Dantzig selector
• Inference for Empirical Wasserstein Distances on Finite Spaces
• 10.1162/jmlr.2003.3.4-5.993
• On Choosing and Bounding Probability Metrics
• Multisample estimation of bacterial composition matrices in metagenomics data
• A New Proof of the Pearson-Fisher Theorem
• Using mixture models for collaborative filtering