{"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/deep-temporal-clustering-fully-unsupervised","title":"Deep Temporal Clustering : Fully Unsupervised Learning of Time-Domain Features","arxiv_id":"1802.01059","date":"2018-02-04","proceeding":null,"authors":["Naveen Sai Madiraju","Seid M. Sadat","Dimitry Fisher","Homa Karimabadi"],"abstract":"Unsupervised learning of time series data, also known as temporal clustering,\nis a challenging problem in machine learning. Here we propose a novel\nalgorithm, Deep Temporal Clustering (DTC), to naturally integrate\ndimensionality reduction and temporal clustering into a single end-to-end\nlearning framework, fully unsupervised. The algorithm utilizes an autoencoder\nfor temporal dimensionality reduction and a novel temporal clustering layer for\ncluster assignment. Then it jointly optimizes the clustering objective and the\ndimensionality reduction objec tive. Based on requirement and application, the\ntemporal clustering layer can be customized with any temporal similarity\nmetric. Several similarity metrics and state-of-the-art algorithms are\nconsidered and compared. To gain insight into temporal features that the\nnetwork has learned for its clustering, we apply a visualization method that\ngenerates a region of interest heatmap for the time series. The viability of\nthe algorithm is demonstrated using time series data from diverse domains,\nranging from earthquakes to spacecraft sensor data. In each case, we show that\nthe proposed algorithm outperforms traditional methods. The superior\nperformance is attributed to the fully integrated temporal dimensionality\nreduction and clustering criterion.","url_abs":"http://arxiv.org/abs/1802.01059v1","url_pdf":"http://arxiv.org/pdf/1802.01059v1.pdf","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":"deep-temporal-clustering-fully-unsupervised","repo_url":"https://github.com/FlorentF9/DeepTemporalClustering","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"deep-temporal-clustering-fully-unsupervised","repo_url":"https://github.com/HamzaG737/Deep-temporal-clustering","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"deep-temporal-clustering-fully-unsupervised","repo_url":"https://github.com/RuiYNU/RATL","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"deep-temporal-clustering-fully-unsupervised","repo_url":"https://github.com/saeeeeru/dtc-tensorflow","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"dimensionality-reduction","task_name":"Dimensionality Reduction"},{"task_slug":"time-series-1","task_name":"Time Series"},{"task_slug":"time-series","task_name":"Time Series Analysis"}],"methods":[{"method_slug":"heatmap","method_name":"Heatmap"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1802.01059","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}