Multi-view self-supervised learning for multivariate variable-channel time series

TL;DR

This paper introduces a self-supervised learning approach for multivariate time series that handles varying input channels and transfers knowledge across datasets, improving performance in biomedical applications.

Contribution

It proposes a novel encoder with message passing neural network for multivariate time series, enabling transfer learning across datasets with different channels.

Findings

Outperforms existing methods with TS2Vec loss.

Effective transfer learning between datasets with different channels.

Pretraining on six channels improves fine-tuning on two channels.

Abstract

Labeling of multivariate biomedical time series data is a laborious and expensive process. Self-supervised contrastive learning alleviates the need for large, labeled datasets through pretraining on unlabeled data. However, for multivariate time series data, the set of input channels often varies between applications, and most existing work does not allow for transfer between datasets with different sets of input channels. We propose learning one encoder to operate on all input channels individually. We then use a message passing neural network to extract a single representation across channels. We demonstrate the potential of this method by pretraining our model on a dataset with six EEG channels and then fine-tuning it on a dataset with two different EEG channels. We compare models with and without the message passing neural network across different contrastive loss functions. We show that our method, combined with the TS2Vec loss, outperforms all other methods in most settings.