{"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/latent-space-autoregression-for-novelty","title":"Latent Space Autoregression for Novelty Detection","arxiv_id":"1807.01653","date":"2018-07-04","proceeding":"CVPR 2019 6","authors":["Davide Abati","Angelo Porrello","Simone Calderara","Rita Cucchiara"],"abstract":"Novelty detection is commonly referred to as the discrimination of\nobservations that do not conform to a learned model of regularity. Despite its\nimportance in different application settings, designing a novelty detector is\nutterly complex due to the unpredictable nature of novelties and its\ninaccessibility during the training procedure, factors which expose the\nunsupervised nature of the problem. In our proposal, we design a general\nframework where we equip a deep autoencoder with a parametric density estimator\nthat learns the probability distribution underlying its latent representations\nthrough an autoregressive procedure. We show that a maximum likelihood\nobjective, optimized in conjunction with the reconstruction of normal samples,\neffectively acts as a regularizer for the task at hand, by minimizing the\ndifferential entropy of the distribution spanned by latent vectors. In addition\nto providing a very general formulation, extensive experiments of our model on\npublicly available datasets deliver on-par or superior performances if compared\nto state-of-the-art methods in one-class and video anomaly detection settings.\nDifferently from prior works, our proposal does not make any assumption about\nthe nature of the novelties, making our work readily applicable to diverse\ncontexts.","url_abs":"http://arxiv.org/abs/1807.01653v2","url_pdf":"http://arxiv.org/pdf/1807.01653v2.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":"latent-space-autoregression-for-novelty","repo_url":"https://github.com/aimagelab/novelty-detection","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"anomaly-detection","task_name":"Anomaly Detection"},{"task_slug":"novelty-detection","task_name":"Novelty Detection"},{"task_slug":"video-anomaly-detection","task_name":"Video Anomaly Detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1807.01653","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1807.01653"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/aimagelab/novelty-detection","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":3},"by_repo_kind":{"official":{"samples":3,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"d30eddff9ca9b668","entry":"normalize","repo":"aimagelab/novelty-detection","repo_kind":"official","path":"utils.py","file_url":"https://github.com/aimagelab/novelty-detection/blob/HEAD/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d30eddff9ca9b668"}},{"code_sha256_prefix":"4ce35e469f1873a4","entry":"novelty_score","repo":"aimagelab/novelty-detection","repo_kind":"official","path":"utils.py","file_url":"https://github.com/aimagelab/novelty-detection/blob/HEAD/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4ce35e469f1873a4"}},{"code_sha256_prefix":"4f27691f34e330ba","entry":"residual_op","repo":"aimagelab/novelty-detection","repo_kind":"official","path":"models/blocks_2d.py","file_url":"https://github.com/aimagelab/novelty-detection/blob/HEAD/models/blocks_2d.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4f27691f34e330ba"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}