{"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-structured-energy-based-models-for","title":"Deep Structured Energy Based Models for Anomaly Detection","arxiv_id":"1605.07717","date":"2016-05-25","proceeding":null,"authors":["Shuangfei Zhai","Yu Cheng","Weining Lu","Zhongfei Zhang"],"abstract":"In this paper, we attack the anomaly detection problem by directly modeling\nthe data distribution with deep architectures. We propose deep structured\nenergy based models (DSEBMs), where the energy function is the output of a\ndeterministic deep neural network with structure. We develop novel model\narchitectures to integrate EBMs with different types of data such as static\ndata, sequential data, and spatial data, and apply appropriate model\narchitectures to adapt to the data structure. Our training algorithm is built\nupon the recent development of score matching \\cite{sm}, which connects an EBM\nwith a regularized autoencoder, eliminating the need for complicated sampling\nmethod. Statistically sound decision criterion can be derived for anomaly\ndetection purpose from the perspective of the energy landscape of the data\ndistribution. We investigate two decision criteria for performing anomaly\ndetection: the energy score and the reconstruction error. Extensive empirical\nstudies on benchmark tasks demonstrate that our proposed model consistently\nmatches or outperforms all the competing methods.","url_abs":"http://arxiv.org/abs/1605.07717v2","url_pdf":"http://arxiv.org/pdf/1605.07717v2.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-structured-energy-based-models-for","repo_url":"https://github.com/intrudetection/robevalanodetect","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"deep-structured-energy-based-models-for","repo_url":"https://github.com/zehuichen123/DSEBM","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"anomaly-detection","task_name":"Anomaly Detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1605.07717","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}