Factorized Topic Models

2013/01/15 by Cheng Zhang, Carl Henrik Ek, Zhang, Cheng +6
Computer Science · Decision Sciences · Social Sciences · #Computer Vision and Pattern Recognition (cs.CV) #Data Quality and Management #FOS: Computer and information sciences #Human Mobility and Location-Based Analysis #Information Retrieval (cs.IR) #Machine Learning (cs.LG) #Topic Modeling #cs.CV #cs.IR #cs.LG

paper · pdf · doi:10.48550/arxiv.1301.3461

ICLR 2013

openalex publication_date 2013/01/15 · arxiv created 2013/04/23 · arxiv updated 2013/04/24 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

In this paper we present a modification to a latent topic model, which makes the model exploit supervision to produce a factorized representation of the observed data. The structured parameterization separately encodes variance that is shared between classes from variance that is private to each class by the introduction of a new prior over the topic space. The approach allows for a more efficient inference and provides an intuitive interpretation of the data in terms of an informative signal together with structured noise. The factorized representation is shown to enhance inference performance for image, text, and video classification.

Citations

• Probabilistic Latent Semantic Analysis
• Distinctive Image Features from Scale-Invariant Keypoints
• Probabilistic topic models
• The Discrete Infinite Logistic Normal Distribution

Related

• Inter-Battery Topic Representation Learning
• Nonparametric Topic Modeling with Neural Inference
• Learning Topic Models - Going beyond SVD
• Supervised Topic Models
• Generalized Topic Modeling