Learning Features with Structure-Adapting Multi-view Exponential Family Harmoniums

TL;DR

This paper introduces SA-MVH, a flexible graphical model for multi-view feature extraction that adapts its structure during training, improving data representation over existing methods.

Contribution

The paper presents a novel structure-adapting multi-view harmonium model with learnable switch parameters controlling view connections, enhancing feature extraction.

Findings

SA-MVH outperforms existing methods on synthetic data

SA-MVH demonstrates improved data representation on real-world datasets

The model adaptively learns structure during training

Abstract

We proposea graphical model for multi-view feature extraction that automatically adapts its structure to achieve better representation of data distribution. The proposed model, structure-adapting multi-view harmonium (SA-MVH) has switch parameters that control the connection between hidden nodes and input views, and learn the switch parameter while training. Numerical experiments on synthetic and a real-world dataset demonstrate the useful behavior of the SA-MVH, compared to existing multi-view feature extraction methods.