{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/deep-autoencoding-gaussian-mixture-model-for","title":"Deep Autoencoding Gaussian Mixture Model for Unsupervised Anomaly Detection","arxiv_id":null,"date":"2018-01-01","proceeding":"ICLR 2018 1","authors":["Bo Zong","Qi Song","Martin Renqiang Min","Wei Cheng","Cristian Lumezanu","Daeki Cho","Haifeng Chen"],"abstract":"Unsupervised anomaly detection on multi- or high-dimensional data is of great importance in both fundamental machine learning research and industrial applications, for which density estimation lies at the core. Although previous approaches based on dimensionality reduction followed by density estimation have made fruitful progress, they mainly suffer from decoupled model learning with inconsistent optimization goals and incapability of preserving essential information in the low-dimensional space. In this paper, we present a Deep Autoencoding Gaussian Mixture Model (DAGMM) for unsupervised anomaly detection. Our model utilizes a deep autoencoder to generate a low-dimensional representation and reconstruction error for each input data point, which is further fed into a Gaussian Mixture Model (GMM). Instead of using decoupled two-stage training and the standard Expectation-Maximization (EM) algorithm, DAGMM jointly optimizes the parameters of the deep autoencoder and the mixture model simultaneously in an end-to-end fashion, leveraging a separate estimation network to facilitate the parameter learning of the mixture model. The joint optimization, which well balances autoencoding reconstruction, density estimation of latent representation, and regularization, helps the autoencoder escape from less attractive local optima and further reduce reconstruction errors, avoiding the need of pre-training. Experimental results on several public benchmark datasets show that, DAGMM significantly outperforms state-of-the-art anomaly detection techniques, and achieves up to 14% improvement based on the standard F1 score.","url_abs":"https://openreview.net/forum?id=BJJLHbb0-","url_pdf":"https://openreview.net/pdf?id=BJJLHbb0-","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-autoencoding-gaussian-mixture-model-for","repo_url":"https://github.com/RomainSabathe/dagmm","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"deep-autoencoding-gaussian-mixture-model-for","repo_url":"https://github.com/danieltan07/dagmm","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"anomaly-detection","task_name":"Anomaly Detection"},{"task_slug":"density-estimation","task_name":"Density Estimation"},{"task_slug":"dimensionality-reduction","task_name":"Dimensionality Reduction"},{"task_slug":"unsupervised-anomaly-detection","task_name":"Unsupervised Anomaly Detection"},{"task_slug":"unsupervised-anomaly-detection-with-specified-6","task_name":"Unsupervised Anomaly Detection with Specified Settings -- 0.1% anomaly"},{"task_slug":"unsupervised-anomaly-detection-with-specified-5","task_name":"Unsupervised Anomaly Detection with Specified Settings -- 1% anomaly"},{"task_slug":"unsupervised-anomaly-detection-with-specified-7","task_name":"Unsupervised Anomaly Detection with Specified Settings -- 10% anomaly"},{"task_slug":"unsupervised-anomaly-detection-with-specified-4","task_name":"Unsupervised Anomaly Detection with Specified Settings -- 20% anomaly"},{"task_slug":"unsupervised-anomaly-detection-with-specified","task_name":"Unsupervised Anomaly Detection with Specified Settings -- 30% anomaly"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-9","task":"Unsupervised Anomaly Detection with Specified Settings -- 0.1% anomaly","dataset":"CIFAR-10","model":"DAGMM","rank_in_archive_order":6,"of":6,"metrics":{"AUC-ROC":"0.494"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-12","task":"Unsupervised Anomaly Detection with Specified Settings -- 0.1% anomaly","dataset":"Cats and Dogs","model":"DAGMM","rank_in_archive_order":4,"of":6,"metrics":{"AUC-ROC":"0.784"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-11","task":"Unsupervised Anomaly Detection with Specified Settings -- 0.1% anomaly","dataset":"Fashion-MNIST","model":"DAGMM","rank_in_archive_order":4,"of":5,"metrics":{"AUC-ROC":"0.784"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-10","task":"Unsupervised Anomaly Detection with Specified Settings -- 0.1% anomaly","dataset":"MNIST","model":"DAGMM","rank_in_archive_order":4,"of":5,"metrics":{"AUC-ROC":"0.624"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-8","task":"Unsupervised Anomaly Detection with Specified Settings -- 0.1% anomaly","dataset":"STL-10","model":"DAGMM","rank_in_archive_order":5,"of":6,"metrics":{"AUC-ROC":"0.574"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-16","task":"Unsupervised Anomaly Detection with Specified Settings -- 1% anomaly","dataset":"CIFAR-10","model":"DAGMM","rank_in_archive_order":6,"of":6,"metrics":{"AUC-ROC":"0.503"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-19","task":"Unsupervised Anomaly Detection with Specified Settings -- 1% anomaly","dataset":"Cats and Dogs","model":"DAGMM","rank_in_archive_order":5,"of":5,"metrics":{"AUC-ROC":"0.710"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-18","task":"Unsupervised Anomaly Detection with Specified Settings -- 1% anomaly","dataset":"Fashion-MNIST","model":"DAGMM","rank_in_archive_order":5,"of":5,"metrics":{"AUC-ROC":"0.793"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-17","task":"Unsupervised Anomaly Detection with Specified Settings -- 1% anomaly","dataset":"MNIST","model":"DAGMM","rank_in_archive_order":5,"of":5,"metrics":{"AUC-ROC":"0.708"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-15","task":"Unsupervised Anomaly Detection with Specified Settings -- 1% anomaly","dataset":"STL-10","model":"DAGMM","rank_in_archive_order":6,"of":6,"metrics":{"AUC-ROC":"0.477"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-21","task":"Unsupervised Anomaly Detection with Specified Settings -- 10% anomaly","dataset":"CIFAR-10","model":"DAGMM","rank_in_archive_order":6,"of":6,"metrics":{"AUC-ROC":"0.778"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-26","task":"Unsupervised Anomaly Detection with Specified Settings -- 10% anomaly","dataset":"Cats and Dogs","model":"DAGMM","rank_in_archive_order":3,"of":6,"metrics":{"AUC-ROC":"0.960"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-27","task":"Unsupervised Anomaly Detection with Specified Settings -- 10% anomaly","dataset":"Fashion-MNIST","model":"DAGMM","rank_in_archive_order":4,"of":5,"metrics":{"AUC-ROC":"0.788"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-22","task":"Unsupervised Anomaly Detection with Specified Settings -- 10% anomaly","dataset":"MNIST","model":"DAGMM","rank_in_archive_order":5,"of":5,"metrics":{"AUC-ROC":"0.629"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-20","task":"Unsupervised Anomaly Detection with Specified Settings -- 10% anomaly","dataset":"STL-10","model":"DAGMM","rank_in_archive_order":5,"of":6,"metrics":{"AUC-ROC":"0.826"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-24","task":"Unsupervised Anomaly Detection with Specified Settings -- 20% anomaly","dataset":"Cats and Dogs","model":"DAGMM","rank_in_archive_order":3,"of":6,"metrics":{"AUC-ROC":"0.914"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-14","task":"Unsupervised Anomaly Detection with Specified Settings -- 20% anomaly","dataset":"Fashion-MNIST","model":"DAGMM","rank_in_archive_order":4,"of":5,"metrics":{"AUC-ROC":"0.780"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-13","task":"Unsupervised Anomaly Detection with Specified Settings -- 20% anomaly","dataset":"MNIST","model":"DAGMM","rank_in_archive_order":5,"of":5,"metrics":{"AUC-ROC":"0.616"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-5","task":"Unsupervised Anomaly Detection with Specified Settings -- 20% anomaly","dataset":"STL-10","model":"DAGMM","rank_in_archive_order":4,"of":6,"metrics":{"AUC-ROC":"0.911"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-6","task":"Unsupervised Anomaly Detection with Specified Settings -- 20% anomaly","dataset":"cifar10","model":"DAGMM","rank_in_archive_order":2,"of":6,"metrics":{"AUC-ROC":"0.883"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-1","task":"Unsupervised Anomaly Detection with Specified Settings -- 30% anomaly","dataset":"ASSIRA Cat Vs Dog","model":"DAGMM","rank_in_archive_order":1,"of":6,"metrics":{"AUC-ROC":"0.846"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-7","task":"Unsupervised Anomaly Detection with Specified Settings -- 30% anomaly","dataset":"CIFAR-10","model":"DAGMM","rank_in_archive_order":2,"of":6,"metrics":{"AUC-ROC":"0.850"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-25","task":"Unsupervised Anomaly Detection with Specified Settings -- 30% anomaly","dataset":"Fashion-MNIST","model":"DAGMM","rank_in_archive_order":4,"of":5,"metrics":{"AUC-ROC":"0.769"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified-23","task":"Unsupervised Anomaly Detection with Specified Settings -- 30% anomaly","dataset":"MNIST","model":"DAGMM","rank_in_archive_order":5,"of":5,"metrics":{"AUC-ROC":"0.613"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-anomaly-detection-with-specified","task":"Unsupervised Anomaly Detection with Specified Settings -- 30% anomaly","dataset":"STL-10","model":"DAGMM","rank_in_archive_order":4,"of":6,"metrics":{"AUC-ROC":"0.883"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers","held":null,"reason":"no_arxiv_id","checked_against":null,"by_repo":[{"repo":"RomainSabathe/dagmm","kind":"listed","samples":0,"ran":0,"instrument":0,"unverified":0},{"repo":"danieltan07/dagmm","kind":"listed","samples":0,"ran":0,"instrument":0,"unverified":0}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"fdd13420d9d83622b9e38cb98a4f998ee9f1f4604cff65806ec3b1dfb54ddde2","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}