Learning Centre

Improving Cross-Subject Activity Recognition via Adversarial Learning

 |  Login

Show simple item record

dc.contributor Aalto-yliopisto fi
dc.contributor Aalto University en
dc.contributor.author Souza Leite, Clayton
dc.contributor.author Xiao, Yu
dc.date.accessioned 2020-06-25T08:40:38Z
dc.date.available 2020-06-25T08:40:38Z
dc.date.issued 2020-05-11
dc.identifier.citation Souza Leite , C & Xiao , Y 2020 , ' Improving Cross-Subject Activity Recognition via Adversarial Learning ' , IEEE Access , vol. 8 , 9091200 , pp. 90542-90554 . https://doi.org/10.1109/ACCESS.2020.2993818 en
dc.identifier.issn 2169-3536
dc.identifier.other PURE UUID: 9b0282e2-c071-48ee-9d1c-b701118273ef
dc.identifier.other PURE ITEMURL: https://research.aalto.fi/en/publications/9b0282e2-c071-48ee-9d1c-b701118273ef
dc.identifier.other PURE LINK: http://www.scopus.com/inward/record.url?scp=85085573859&partnerID=8YFLogxK
dc.identifier.other PURE FILEURL: https://research.aalto.fi/files/43528914/Leite_Improving_cross_subject_IEEEAccess.pdf
dc.identifier.uri https://aaltodoc.aalto.fi/handle/123456789/45177
dc.description | openaire: EC/H2020/777222/EU//ATTRACT
dc.description.abstract Deep learning has been widely used for implementing human activity recognition from wearable sensors like inertial measurement units. The performance of deep activity recognition is heavily affected by the amount and variability of the labeled data available for training the deep learning models. On the other hand, it is costly and time-consuming to collect and label data. Given limited training data, it is hard to maintain high performance across a wide range of subjects, due to the differences in the underlying data distribution of the training and the testing sets. In this work, we develop a novel solution that applies adversarial learning to improve cross-subject performance by generating training data that mimic artificial subjects - i.e. through data augmentation - and enforcing the activity classifier to ignore subject-dependent information. Contrary to domain adaptation methods, our solution does not utilize any data from subjects of the test set (or target domain). Furthermore, our solution is versatile as it can be utilized together with any deep neural network as the classifier. Considering the open dataset PAMAP2, nearly 10% higher cross-subject performance in terms of F1-score can be achieved when training a CNN-LSTM-based classifier with our solution. A performance gain of 5% is also observed when our solution is applied to a state-of-the-art HAR classifier composed of a combination of inception neural network and recurrent neural network. We also investigate different influencing factors of classification performance (i.e. selection of sensor modalities, sampling rates and the number of subjects in the training data), and summarize a practical guideline for implementing deep learning solutions for sensor-based human activity recognition. en
dc.format.extent 13
dc.format.extent 90542-90554
dc.format.mimetype application/pdf
dc.language.iso en en
dc.relation info:eu-repo/grantAgreement/EC/H2020/777222/EU//ATTRACT
dc.relation.ispartofseries IEEE Access en
dc.relation.ispartofseries Volume 8 en
dc.rights openAccess en
dc.title Improving Cross-Subject Activity Recognition via Adversarial Learning en
dc.type A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä fi
dc.description.version Peer reviewed en
dc.contributor.department Department of Communications and Networking
dc.subject.keyword Human activity recognition
dc.subject.keyword deep learning
dc.subject.keyword adversarial learning
dc.subject.keyword data augmentation
dc.subject.keyword cross-subject performance
dc.identifier.urn URN:NBN:fi:aalto-202006254134
dc.identifier.doi 10.1109/ACCESS.2020.2993818
dc.type.version publishedVersion

Files in this item

Files Size Format View

There are no open access files associated with this item.

This item appears in the following Collection(s)

Show simple item record

Search archive

Advanced Search

article-iconSubmit a publication