Federated Self-Supervised Learning of Multisensor Representations for Embedded Intelligence

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 scalogram-signal correspondence learning based on wavelet transform (WT) to learn useful representations from unlabeled sensor inputs 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 view (i.e., a scalogram generated with WT) align with each other, by optimizing a contrastive objective. We extensively assess the quality of learned features with our multiview strategy on diverse public data sets, 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 works significantly better than pretraining with autoencoders in both central and federated contexts. Notably, it improves the generalization in a semisupervised setting as it reduces the volume of labeled data required through leveraging self-supervised learning.