{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/leveraging-label-information-in-a-knowledge","title":"Leveraging Label Information in a Knowledge-Driven Approach for Rolling-Element Bearings Remaining Useful Life Prediction","arxiv_id":null,"date":"2021-04-14","proceeding":null,"authors":["Tarek Berghout","Mohamed Benbouzid","Leïla-Hayet Mouss"],"abstract":"Since bearing deterioration patterns are difficult to collect from real, long lifetime scenarios,\r\ndata-driven research has been directed towards recovering them by imposing accelerated life tests.\r\nConsequently, insufficiently recovered features due to rapid damage propagation seem more likely\r\nto lead to poorly generalized learning machines. Knowledge-driven learning comes as a solution by\r\nproviding prior assumptions from transfer learning. Likewise, the absence of true labels was able to\r\ncreate inconsistency related problems between samples, and teacher-given label behaviors led to more\r\nill-posed predictors. Therefore, in an attempt to overcome the incomplete, unlabeled data drawbacks,\r\na new autoencoder has been designed as an additional source that could correlate inputs and labels\r\nby exploiting label information in a completely unsupervised learning scheme. Additionally, its\r\nstacked denoising version seems to more robustly be able to recover them for new unseen data. Due\r\nto the non-stationary and sequentially driven nature of samples, recovered representations have\r\nbeen fed into a transfer learning, convolutional, long–short-term memory neural network for further\r\nmeaningful learning representations. The assessment procedures were benchmarked against recent\r\nmethods under different training datasets. The obtained results led to more efficiency confirming the\r\nstrength of the new learning path.","url_abs":"https://www.mdpi.com/1996-1073/14/8/2163","url_pdf":"https://www.mdpi.com/1996-1073/14/8/2163","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"leveraging-label-information-in-a-knowledge","repo_url":"https://github.com/TBdevellopper/MY_MATLAB_codes/blob/a1d14518150275fad9b5626a938b8e94430d74ef/Basic_version_of_the_proposed%20autoencouder.zip","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"denoising","task_name":"Denoising"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}