arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2102.06073cs.LG
收藏

SelfHAR:通过使用未标记数据进行自训练来改进人类活动识别

SelfHAR: Improving Human Activity Recognition through Self-training with Unlabeled Data

University of Cambridge(剑桥大学);The Alan Turing Institute(艾伦·图灵研究所)

Chi Ian Tang, Ignacio Perez-Pozuelo, Dimitris Spathis, Soren Brage, Nick Wareham, Cecilia Mascolo

摘要:

机器学习和深度学习在包括人类活动识别在内的移动传感应用中显示出巨大的前景。然而,这些模型在现实环境中的性能在很大程度上取决于捕获不同行为的大型数据集的可用性。最近,计算机视觉和自然语言处理的研究表明,利用大量未标记数据可以实现与最先进的监督模型相当的性能。在这项工作中,我们提出了SelfHAR,这是一种半监督模型,可以有效地学习利用未标记的移动传感数据集来补充小的标记数据集。我们的方法结合了师生自我训练,它提取了未标记和标记数据集的知识,同时允许数据增强,以及多任务自我监督,通过预测输入的失真版本来学习鲁棒的信号电平表示。我们在各种HAR数据集上评估了SelfHAR,并显示了监督和以前的半监督方法的最新性能,在推理时使用相同数量的模型参数,F1得分增加了12%。此外,SelfHAR是数据高效的,与监督方法相比,使用多达10倍的标记数据达到类似的性能。我们的工作不仅在不同的HAR数据集中实现了最先进的性能,而且还揭示了预训练任务如何影响下游性能。

英文摘要:

Machine learning and deep learning have shown great promise in mobile sensing applications, including Human Activity Recognition. However, the performance of such models in real-world settings largely depends on the availability of large datasets that captures diverse behaviors. Recently, studies in computer vision and natural language processing have shown that leveraging massive amounts of unlabeled data enables performance on par with state-of-the-art supervised models. In this work, we present SelfHAR, a semi-supervised model that effectively learns to leverage unlabeled mobile sensing datasets to complement small labeled datasets. Our approach combines teacher-student self-training, which distills the knowledge of unlabeled and labeled datasets while allowing for data augmentation, and multi-task self-supervision, which learns robust signal-level representations by predicting distorted versions of the input. We evaluated SelfHAR on various HAR datasets and showed state-of-the-art performance over supervised and previous semi-supervised approaches, with up to 12% increase in F1 score using the same number of model parameters at inference. Furthermore, SelfHAR is data-efficient, reaching similar performance using up to 10 times less labeled data compared to supervised approaches. Our work not only achieves state-of-the-art performance in a diverse set of HAR datasets, but also sheds light on how pre-training tasks may affect downstream performance.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑