Mixed-Variate Restricted Boltzmann Machines

TL;DR

This paper introduces Mixed-Variate Restricted Boltzmann Machines, a model capable of handling heterogeneous data types and modalities for tasks like feature extraction, classification, and data completion, demonstrated on a large-scale survey dataset.

Contribution

The paper proposes a novel RBM variant that models multiple variable types simultaneously, enabling versatile applications in heterogeneous data analysis.

Findings

Effective feature extraction and visualization from survey data

Accurate data completion and prediction results

Versatile use as a pre-processing, classification, and data imputation tool

Abstract

Modern datasets are becoming heterogeneous. To this end, we present in this paper Mixed-Variate Restricted Boltzmann Machines for simultaneously modelling variables of multiple types and modalities, including binary and continuous responses, categorical options, multicategorical choices, ordinal assessment and category-ranked preferences. Dependency among variables is modeled using latent binary variables, each of which can be interpreted as a particular hidden aspect of the data. The proposed model, similar to the standard RBMs, allows fast evaluation of the posterior for the latent variables. Hence, it is naturally suitable for many common tasks including, but not limited to, (a) as a pre-processing step to convert complex input data into a more convenient vectorial representation through the latent posteriors, thereby offering a dimensionality reduction capacity, (b) as a classifier supporting binary, multiclass, multilabel, and label-ranking outputs, or a regression tool for continuous outputs and (c) as a data completion tool for multimodal and heterogeneous data. We evaluate the proposed model on a large-scale dataset using the world opinion survey results on three tasks: feature extraction and visualization, data completion and prediction.