Transformer-Based Self-Supervised Learning for Emotion Recognition

TL;DR

This paper introduces a Transformer-based self-supervised learning approach for emotion recognition from ECG signals, achieving state-of-the-art results by leveraging large unlabeled datasets for pre-training.

Contribution

It presents a novel application of Transformers and self-supervised pre-training to physiological signal-based emotion recognition, improving performance on ECG data.

Findings

Transformer-based models outperform previous methods.

Self-supervised pre-training enhances emotion recognition accuracy.

State-of-the-art results achieved on the AMIGOS dataset.

Abstract

In order to exploit representations of time-series signals, such as physiological signals, it is essential that these representations capture relevant information from the whole signal. In this work, we propose to use a Transformer-based model to process electrocardiograms (ECG) for emotion recognition. Attention mechanisms of the Transformer can be used to build contextualized representations for a signal, giving more importance to relevant parts. These representations may then be processed with a fully-connected network to predict emotions. To overcome the relatively small size of datasets with emotional labels, we employ self-supervised learning. We gathered several ECG datasets with no labels of emotion to pre-train our model, which we then fine-tuned for emotion recognition on the AMIGOS dataset. We show that our approach reaches state-of-the-art performances for emotion recognition using ECG signals on AMIGOS. More generally, our experiments show that transformers and pre-training are promising strategies for emotion recognition with physiological signals.