Topic Modeling in Embedding Spaces

• Published in: Transactions of the Association for Computational Linguistics Vol. 8; pp. 439 - 453
• Main Authors: Dieng, Adji B., Ruiz, Francisco J. R., Blei, David M.
• Format: Journal Article
• Language: English
• Published: One Rogers Street, Cambridge, MA 02142-1209, USA MIT Press 01.01.2020; MIT Press Journals, The; The MIT Press
• Subjects: Algorithms; Computational linguistics; Data mining; Datasets; Dirichlet problem; Documents; Embedding; Inference; Linguistics; Modelling; Performance prediction; Semantics; Topics
• ISSN: 2307-387X; 2307-387X
• DOI: 10.1162/tacl_a_00325

Topic modeling analyzes documents to learn meaningful patterns of words. However, existing topic models fail to learn interpretable topics when working with large and heavy-tailed vocabularies. To this end, we develop the ( ), a generative model of documents that marries traditional topic models with word embeddings. More specifically, the models each word with a categorical distribution whose natural parameter is the inner product between the word’s embedding and an embedding of its assigned topic. To fit the , we develop an efficient amortized variational inference algorithm. The discovers interpretable topics even with large vocabularies that include rare words and stop words. It outperforms existing document models, such as latent Dirichlet allocation, in terms of both topic quality and predictive performance.