{"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/transformation-autoregressive-networks","title":"Transformation Autoregressive Networks","arxiv_id":"1801.09819","date":"2018-01-30","proceeding":"ICML 2018 7","authors":["Junier B. Oliva","Avinava Dubey","Manzil Zaheer","Barnabás Póczos","Ruslan Salakhutdinov","Eric P. Xing","Jeff Schneider"],"abstract":"The fundamental task of general density estimation $p(x)$ has been of keen\ninterest to machine learning. In this work, we attempt to systematically\ncharacterize methods for density estimation. Broadly speaking, most of the\nexisting methods can be categorized into either using: \\textit{a})\nautoregressive models to estimate the conditional factors of the chain rule,\n$p(x_{i}\\, |\\, x_{i-1}, \\ldots)$; or \\textit{b}) non-linear transformations of\nvariables of a simple base distribution. Based on the study of the\ncharacteristics of these categories, we propose multiple novel methods for each\ncategory. For example we proposed RNN based transformations to model\nnon-Markovian dependencies. Further, through a comprehensive study over both\nreal world and synthetic data, we show for that jointly leveraging\ntransformations of variables and autoregressive conditional models, results in\na considerable improvement in performance. We illustrate the use of our models\nin outlier detection and image modeling. Finally we introduce a novel data\ndriven framework for learning a family of distributions.","url_abs":"http://arxiv.org/abs/1801.09819v5","url_pdf":"http://arxiv.org/pdf/1801.09819v5.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":[],"tasks":[{"task_slug":"density-estimation","task_name":"Density Estimation"},{"task_slug":"outlier-detection","task_name":"Outlier Detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/density-estimation-on-bsds300","task":"Density Estimation","dataset":"BSDS300","model":"TAN","rank_in_archive_order":1,"of":5,"metrics":{"Log-likelihood":"159.8"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1801.09819","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}