ColloSSL: Collaborative Self-Supervised Learning for Human Activity Recognition

TL;DR

ColloSSL introduces a novel multi-device self-supervised learning method for human activity recognition, leveraging unlabeled multi-device sensor data to improve feature learning and outperform supervised methods, especially with limited labeled data.

Contribution

The paper proposes ColloSSL, a multi-device self-supervised learning framework with new device selection, sampling, and a multi-view contrastive loss, advancing HAR without extensive labeled data.

Findings

Outperforms supervised and semi-supervised methods on multiple datasets.

Achieves up to 7.9% higher F1 score than baselines.

Effective in low-data regimes, using only one-tenth of labeled data.

Abstract

A major bottleneck in training robust Human-Activity Recognition models (HAR) is the need for large-scale labeled sensor datasets. Because labeling large amounts of sensor data is an expensive task, unsupervised and semi-supervised learning techniques have emerged that can learn good features from the data without requiring any labels. In this paper, we extend this line of research and present a novel technique called Collaborative Self-Supervised Learning (ColloSSL) which leverages unlabeled data collected from multiple devices worn by a user to learn high-quality features of the data. A key insight that underpins the design of ColloSSL is that unlabeled sensor datasets simultaneously captured by multiple devices can be viewed as natural transformations of each other, and leveraged to generate a supervisory signal for representation learning. We present three technical innovations to extend conventional self-supervised learning algorithms to a multi-device setting: a Device Selection approach which selects positive and negative devices to enable contrastive learning, a Contrastive Sampling algorithm which samples positive and negative examples in a multi-device setting, and a loss function called Multi-view Contrastive Loss which extends standard contrastive loss to a multi-device setting. Our experimental results on three multi-device datasets show that ColloSSL outperforms both fully-supervised and semi-supervised learning techniques in majority of the experiment settings, resulting in an absolute increase of upto 7.9% in F_1 score compared to the best performing baselines. We also show that ColloSSL outperforms the fully-supervised methods in a low-data regime, by just using one-tenth of the available labeled data in the best case.