Multimodal Data Fusion based on the Global Workspace Theory

TL;DR

This paper introduces the Global Workspace Network, a neural architecture inspired by cognitive science, that improves multimodal data fusion by effectively handling uncertainties and outperforming traditional methods in pain classification tasks.

Contribution

The paper presents the GWN, a novel neural network model inspired by the Global Workspace Theory, for dynamic multimodal data fusion with improved uncertainty handling.

Findings

GWN achieved an average F1 score of 0.92 in pain discrimination.

GWN achieved an average F1 score of 0.75 in pain level classification.

GWN significantly outperforms traditional concatenation-based fusion methods.

Abstract

We propose a novel neural network architecture, named the Global Workspace Network (GWN), which addresses the challenge of dynamic and unspecified uncertainties in multimodal data fusion. Our GWN is a model of attention across modalities and evolving through time, and is inspired by the well-established Global Workspace Theory from the field of cognitive science. The GWN achieved average F1 score of 0.92 for discrimination between pain patients and healthy participants and average F1 score = 0.75 for further classification of three pain levels for a patient, both based on the multimodal EmoPain dataset captured from people with chronic pain and healthy people performing different types of exercise movements in unconstrained settings. In these tasks, the GWN significantly outperforms the typical fusion approach of merging by concatenation. We further provide extensive analysis of the behaviour of the GWN and its ability to address uncertainties (hidden noise) in multimodal data.