{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T14:40:26Z","timestamp":1777560026653,"version":"3.51.4"},"reference-count":29,"publisher":"SAGE Publications","issue":"4","license":[{"start":{"date-parts":[[2023,10,13]],"date-time":"2023-10-13T00:00:00Z","timestamp":1697155200000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["AIC"],"published-print":{"date-parts":[[2023,10,13]]},"abstract":"<jats:p>As the research community focuses on improving the reliability of deep learning, identifying out-of-distribution (OOD) data has become crucial. Detecting OOD inputs during test\/prediction allows the model to account for discriminative features unknown to the model. This capability increases the model\u2019s reliability since this model provides a class prediction solely at incoming data similar to the training one. Although OOD detection is well-established in computer vision, it is relatively unexplored in other areas, like time series-based human activity recognition (HAR). Since uncertainty has been a critical driver for OOD in vision-based models, the same component has proven effective in time-series applications. In this work, we propose an ensemble-based temporal learning framework to address the OOD detection problem in HAR with time-series data. First, we define different types of OOD for HAR that arise from realistic scenarios. Then we apply our ensemble-based temporal learning framework incorporating uncertainty to detect OODs for the defined HAR workloads. This particular formulation also allows a novel approach to fall detection. We train our model on non-fall activities and detect falls as OOD. Our method shows state-of-the-art performance in a fall detection task using much lesser data. Furthermore, the ensemble framework outperformed the traditional deep-learning method (our baseline) on the OOD detection task across all the other chosen datasets.<\/jats:p>","DOI":"10.3233\/aic-220205","type":"journal-article","created":{"date-parts":[[2023,8,29]],"date-time":"2023-08-29T11:13:53Z","timestamp":1693307633000},"page":"251-267","source":"Crossref","is-referenced-by-count":1,"title":["Classifying falls using out-of-distribution detection in human activity recognition"],"prefix":"10.1177","volume":"36","author":[{"given":"Debaditya","family":"Roy","sequence":"first","affiliation":[{"name":"Department of Electrical Engineering and Computer Science (EECS), Royal Institute of Technology (KTH), Stockholm, Sweden"},{"name":"Qamcom Research and Technology, Stockholm, Sweden"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Vangjush","family":"Komini","sequence":"additional","affiliation":[{"name":"Department of Electrical Engineering and Computer Science (EECS), Royal Institute of Technology (KTH), Stockholm, Sweden"},{"name":"Qamcom Research and Technology, Stockholm, Sweden"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sarunas","family":"Girdzijauskas","sequence":"additional","affiliation":[{"name":"Department of Electrical Engineering and Computer Science (EECS), Royal Institute of Technology (KTH), Stockholm, Sweden"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"179","reference":[{"key":"10.3233\/AIC-220205_ref1","doi-asserted-by":"crossref","unstructured":"A.\u00a0Akbari and R.\u00a0Jafari, A deep learning assisted method for measuring uncertainty in activity recognition with wearable sensors, in: 2019 IEEE EMBS International Conference on Biomedical & Health Informatics (BHI), IEEE, 2019, pp.\u00a01\u20135.","DOI":"10.1109\/BHI.2019.8834505"},{"key":"10.3233\/AIC-220205_ref2","doi-asserted-by":"publisher","DOI":"10.1145\/3302506.3310391"},{"key":"10.3233\/AIC-220205_ref3","unstructured":"D.\u00a0Anguita, A.\u00a0Ghio, L.\u00a0Oneto, X.\u00a0Parra Perez and J.L.\u00a0Reyes Ortiz, A public domain dataset for human activity recognition using smartphones, in: Proceedings of the 21st International European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, 2013, pp.\u00a0437\u2013442."},{"key":"10.3233\/AIC-220205_ref4","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-21671-3_9"},{"key":"10.3233\/AIC-220205_ref5","doi-asserted-by":"crossref","unstructured":"L.\u00a0Bao and S.S.\u00a0Intille, Activity recognition from user-annotated acceleration data, in: International Conference on Pervasive Computing, Springer, 2004, pp.\u00a01\u201317.","DOI":"10.1007\/978-3-540-24646-6_1"},{"key":"10.3233\/AIC-220205_ref6","unstructured":"C.\u00a0Blundell, J.\u00a0Cornebise, K.\u00a0Kavukcuoglu and D.\u00a0Wierstra, Weight uncertainty in neural network, in: International Conference on Machine Learning, PMLR, 2015, pp.\u00a01613\u20131622."},{"key":"10.3233\/AIC-220205_ref8","first-page":"7068","article-title":"Exploring the limits of out-of-distribution detection","volume":"34","author":"Fort","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.3233\/AIC-220205_ref9","unstructured":"Y.\u00a0Gal and Z.\u00a0Ghahramani, Dropout as a Bayesian approximation: Representing model uncertainty in deep learning, in: International Conference on Machine Learning, PMLR, 2016, pp.\u00a01050\u20131059."},{"issue":"10","key":"10.3233\/AIC-220205_ref13","doi-asserted-by":"publisher","first-page":"1973","DOI":"10.1109\/TPAMI.2016.2621761","article-title":"Learning to recognize human activities using soft labels","volume":"39","author":"Hu","year":"2016","journal-title":"IEEE transactions on pattern analysis and machine intelligence"},{"key":"10.3233\/AIC-220205_ref14","doi-asserted-by":"publisher","first-page":"915","DOI":"10.1016\/j.asoc.2017.09.027","article-title":"Real-time human activity recognition from accelerometer data using convolutional neural networks","volume":"62","author":"Ignatov","year":"2018","journal-title":"Applied Soft Computing"},{"issue":"01","key":"10.3233\/AIC-220205_ref15","doi-asserted-by":"publisher","first-page":"70","DOI":"10.3414\/ME9103","article-title":"Anomaly detection using temporal data mining in a smart home environment","volume":"47","author":"Jakkula","year":"2008","journal-title":"Methods of information in medicine"},{"key":"10.3233\/AIC-220205_ref16","doi-asserted-by":"publisher","DOI":"10.23919\/FRUCT.2017.8071301"},{"key":"10.3233\/AIC-220205_ref17","unstructured":"A.\u00a0Kendall and Y.\u00a0Gal, What uncertainties do we need in Bayesian deep learning for computer vision?, in: Advances in Neural Information Processing Systems, Vol.\u00a030, 2017."},{"key":"10.3233\/AIC-220205_ref18","unstructured":"D.P.\u00a0Kingma and M.\u00a0Welling, Stochastic gradient VB and the variational auto-encoder, in: Second International Conference on Learning Representations, ICLR, Vol.\u00a019, 2014, p.\u00a0121."},{"key":"10.3233\/AIC-220205_ref21","unstructured":"B.\u00a0Lakshminarayanan, A.\u00a0Pritzel and C.\u00a0Blundell, Simple and scalable predictive uncertainty estimation using deep ensembles, in: Advances in Neural Information Processing Systems, Vol.\u00a030, 2017."},{"key":"10.3233\/AIC-220205_ref22","unstructured":"K.\u00a0Lee, K.\u00a0Lee, H.\u00a0Lee and J.\u00a0Shin, A simple unified framework for detecting out-of-distribution samples and adversarial attacks, in: Advances in Neural Information Processing Systems, Vol.\u00a031, 2018."},{"key":"10.3233\/AIC-220205_ref23","doi-asserted-by":"crossref","unstructured":"S.-M.\u00a0Lee, S.M.\u00a0Yoon and H.\u00a0Cho, Human activity recognition from accelerometer data using convolutional neural network, in: 2017 IEEE International Conference on Big Data and Smart Computing (BigComp), IEEE, 2017, pp.\u00a0131\u2013134.","DOI":"10.1109\/BIGCOMP.2017.7881728"},{"key":"10.3233\/AIC-220205_ref25","doi-asserted-by":"publisher","DOI":"10.1145\/3302505.3310068"},{"issue":"7540","key":"10.3233\/AIC-220205_ref27","doi-asserted-by":"publisher","first-page":"529","DOI":"10.1038\/nature14236","article-title":"Human-level control through deep reinforcement learning","volume":"518","author":"Mnih","year":"2015","journal-title":"nature"},{"key":"10.3233\/AIC-220205_ref28","doi-asserted-by":"publisher","DOI":"10.3390\/s17112556"},{"issue":"1","key":"10.3233\/AIC-220205_ref29","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13673-019-0194-5","article-title":"Multi-sensor fusion based on multiple classifier systems for human activity identification","volume":"9","author":"Nweke","year":"2019","journal-title":"Human-centric Computing and Information Sciences"},{"key":"10.3233\/AIC-220205_ref30","doi-asserted-by":"publisher","DOI":"10.3390\/s16010115"},{"key":"10.3233\/AIC-220205_ref31","doi-asserted-by":"publisher","DOI":"10.3390\/s21196566"},{"key":"10.3233\/AIC-220205_ref32","doi-asserted-by":"publisher","DOI":"10.3390\/s17010198"},{"key":"10.3233\/AIC-220205_ref33","doi-asserted-by":"crossref","unstructured":"A.\u00a0Vyas, N.\u00a0Jammalamadaka, X.\u00a0Zhu, D.\u00a0Das, B.\u00a0Kaul and T.L.\u00a0Willke, Out-of-distribution detection using an ensemble of self supervised leave-out classifiers, in: Proceedings of the European Conference on Computer Vision (ECCV), 2018, pp.\u00a0550\u2013564.","DOI":"10.1007\/978-3-030-01237-3_34"},{"key":"10.3233\/AIC-220205_ref34","doi-asserted-by":"publisher","DOI":"10.3390\/s16020189"},{"key":"10.3233\/AIC-220205_ref35","doi-asserted-by":"publisher","DOI":"10.1109\/BHI.2016.7455925"},{"key":"10.3233\/AIC-220205_ref36","doi-asserted-by":"publisher","first-page":"133190","DOI":"10.1109\/ACCESS.2019.2940729","article-title":"Smartphone and smartwatch-based biometrics using activities of daily living","volume":"7","author":"Weiss","year":"2019","journal-title":"IEEE Access"},{"key":"10.3233\/AIC-220205_ref37","doi-asserted-by":"crossref","unstructured":"M.\u00a0Yamauchi, Y.\u00a0Ohsita, M.\u00a0Murata, K.\u00a0Ueda and Y.\u00a0Kato, Anomaly detection for smart home based on user behavior, in: 2019 IEEE International Conference on Consumer Electronics (ICCE), IEEE, 2019, pp.\u00a01\u20136.","DOI":"10.1109\/ICCE.2019.8661976"}],"container-title":["AI Communications"],"original-title":[],"link":[{"URL":"https:\/\/content.iospress.com\/download?id=10.3233\/AIC-220205","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,28]],"date-time":"2026-04-28T18:28:09Z","timestamp":1777400889000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/full\/10.3233\/AIC-220205"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,10,13]]},"references-count":29,"journal-issue":{"issue":"4"},"URL":"https:\/\/doi.org\/10.3233\/aic-220205","relation":{},"ISSN":["1875-8452","0921-7126"],"issn-type":[{"value":"1875-8452","type":"electronic"},{"value":"0921-7126","type":"print"}],"subject":[],"published":{"date-parts":[[2023,10,13]]}}}