Please use this identifier to cite or link to this item:
https://doi.org/10.1109/access.2021.3067455
DC Field | Value | |
---|---|---|
dc.title | Using Under-Trained Deep Ensembles to Learn under Extreme Label Noise: A Case Study for Sleep Apnea Detection | |
dc.contributor.author | Nikolaidis, Konstantinos | |
dc.contributor.author | Plagemann, Thomas | |
dc.contributor.author | Kristiansen, Stein | |
dc.contributor.author | Goebel, Vera | |
dc.contributor.author | Kankanhalli, Mohan | |
dc.date.accessioned | 2022-10-13T01:16:03Z | |
dc.date.available | 2022-10-13T01:16:03Z | |
dc.date.issued | 2021-01-01 | |
dc.identifier.citation | Nikolaidis, Konstantinos, Plagemann, Thomas, Kristiansen, Stein, Goebel, Vera, Kankanhalli, Mohan (2021-01-01). Using Under-Trained Deep Ensembles to Learn under Extreme Label Noise: A Case Study for Sleep Apnea Detection. IEEE Access 9 : 45919-45934. ScholarBank@NUS Repository. https://doi.org/10.1109/access.2021.3067455 | |
dc.identifier.issn | 2169-3536 | |
dc.identifier.uri | https://scholarbank.nus.edu.sg/handle/10635/232870 | |
dc.description.abstract | Improper or erroneous labelling can pose a hindrance to reliable generalization for supervised learning. This can have negative consequences, especially for critical fields such as healthcare. We propose an effective new approach for learning under extreme label noise for medical applications like sleep apnea, that is based on under-trained deep ensembles. Each ensemble member is trained with a subset of the training data, to acquire a general overview of the decision boundary separation, without focusing on potentially erroneous details. The accumulated knowledge of the ensemble is combined to form new labels, that determine a better class separation than the original labels. A new model is trained with these labels to generalize reliably despite the label noise. We evaluate our approach on the tasks of sleep apnea detection and sleep apnea severity classification, and observe performance improvement in kappa from 0.02 up-to 0.55. © 2013 IEEE. | |
dc.publisher | Institute of Electrical and Electronics Engineers Inc. | |
dc.rights | Attribution 4.0 International | |
dc.rights.uri | https://creativecommons.org/licenses/by/4.0/ | |
dc.source | Scopus OA2021 | |
dc.subject | Biomedical informatics | |
dc.subject | label noise | |
dc.subject | machine learning | |
dc.subject | sleep apnea | |
dc.subject | supervised learning | |
dc.type | Article | |
dc.contributor.department | DEPARTMENT OF INFORMATION SYSTEMS AND ANALYTICS | |
dc.description.doi | 10.1109/access.2021.3067455 | |
dc.description.sourcetitle | IEEE Access | |
dc.description.volume | 9 | |
dc.description.page | 45919-45934 | |
Appears in Collections: | Elements Staff Publications |
Show simple item record
Files in This Item:
File | Description | Size | Format | Access Settings | Version | |
---|---|---|---|---|---|---|
10_1109_access_2021_3067455.pdf | 1.26 MB | Adobe PDF | OPEN | None | View/Download |
This item is licensed under a Creative Commons License