Factorized Topic Models
This work addresses the need for more interpretable and efficient topic modeling in multi-class classification tasks, though it appears incremental as a modification to existing models.
The authors tackled the problem of improving latent topic models by introducing a factorized representation that separates shared and class-private variance, resulting in enhanced inference performance for image, text, and video classification.
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 eff{}icient 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.