CroSSL: Cross-modal Self-Supervised Learning for Time-series through Latent Masking

TL;DR

CroSSL introduces a novel self-supervised learning framework for multimodal time-series data that employs latent masking and cross-modal aggregation, enabling effective learning even with missing data and minimal labels.

Contribution

The paper proposes CroSSL, a new cross-modal self-supervised learning method that uses latent masking and aggregation, improving robustness and versatility over existing SSL approaches.

Findings

Outperforms previous SSL and supervised benchmarks with minimal labeled data.

Effectively handles missing modalities and data corruption.

Enhances cross-modal learning through latent masking strategies.

Abstract

Limited availability of labeled data for machine learning on multimodal time-series extensively hampers progress in the field. Self-supervised learning (SSL) is a promising approach to learning data representations without relying on labels. However, existing SSL methods require expensive computations of negative pairs and are typically designed for single modalities, which limits their versatility. We introduce CroSSL (Cross-modal SSL), which puts forward two novel concepts: masking intermediate embeddings produced by modality-specific encoders, and their aggregation into a global embedding through a cross-modal aggregator that can be fed to down-stream classifiers. CroSSL allows for handling missing modalities and end-to-end cross-modal learning without requiring prior data preprocessing for handling missing inputs or negative-pair sampling for contrastive learning. We evaluate our method on a wide range of data, including motion sensors such as accelerometers or gyroscopes and biosignals (heart rate, electroencephalograms, electromyograms, electrooculograms, and electrodermal) to investigate the impact of masking ratios and masking strategies for various data types and the robustness of the learned representations to missing data. Overall, CroSSL outperforms previous SSL and supervised benchmarks using minimal labeled data, and also sheds light on how latent masking can improve cross-modal learning. Our code is open-sourced at https://github.com/dr-bell/CroSSL.