Federated Self-Supervised Learning of Multi-Sensor Representations for Embedded Intelligence

TL;DR

This paper introduces a federated self-supervised learning method using wavelet-based scalogram-signal correspondence to learn sensor data representations without labels, enabling effective downstream tasks and reducing labeled data needs.

Contribution

It presents a novel wavelet transform-based self-supervised approach for federated learning on unlabeled multi-sensor data, outperforming autoencoder pre-training and enhancing generalization.

Findings

Achieves strong performance on diverse datasets

Outperforms autoencoder pre-training in federated settings

Reduces labeled data requirements in semi-supervised learning

Abstract

Smartphones, wearables, and Internet of Things (IoT) devices produce a wealth of data that cannot be accumulated in a centralized repository for learning supervised models due to privacy, bandwidth limitations, and the prohibitive cost of annotations. Federated learning provides a compelling framework for learning models from decentralized data, but conventionally, it assumes the availability of labeled samples, whereas on-device data are generally either unlabeled or cannot be annotated readily through user interaction. To address these issues, we propose a self-supervised approach termed \textit{scalogram-signal correspondence learning} based on wavelet transform to learn useful representations from unlabeled sensor inputs, such as electroencephalography, blood volume pulse, accelerometer, and WiFi channel state information. Our auxiliary task requires a deep temporal neural network to determine if a given pair of a signal and its complementary viewpoint (i.e., a scalogram generated with a wavelet transform) align with each other or not through optimizing a contrastive objective. We extensively assess the quality of learned features with our multi-view strategy on diverse public datasets, achieving strong performance in all domains. We demonstrate the effectiveness of representations learned from an unlabeled input collection on downstream tasks with training a linear classifier over pretrained network, usefulness in low-data regime, transfer learning, and cross-validation. Our methodology achieves competitive performance with fully-supervised networks, and it outperforms pre-training with autoencoders in both central and federated contexts. Notably, it improves the generalization in a semi-supervised setting as it reduces the volume of labeled data required through leveraging self-supervised learning.