{"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/convolutional-normalizing-flows","title":"Convolutional Normalizing Flows","arxiv_id":"1711.02255","date":"2017-11-07","proceeding":"ICLR 2018 1","authors":["Guoqing Zheng","Yiming Yang","Jaime Carbonell"],"abstract":"Bayesian posterior inference is prevalent in various machine learning\nproblems. Variational inference provides one way to approximate the posterior\ndistribution, however its expressive power is limited and so is the accuracy of\nresulting approximation. Recently, there has a trend of using neural networks\nto approximate the variational posterior distribution due to the flexibility of\nneural network architecture. One way to construct flexible variational\ndistribution is to warp a simple density into a complex by normalizing flows,\nwhere the resulting density can be analytically evaluated. However, there is a\ntrade-off between the flexibility of normalizing flow and computation cost for\nefficient transformation. In this paper, we propose a simple yet effective\narchitecture of normalizing flows, ConvFlow, based on convolution over the\ndimensions of random input vector. Experiments on synthetic and real world\nposterior inference problems demonstrate the effectiveness and efficiency of\nthe proposed method.","url_abs":"http://arxiv.org/abs/1711.02255v2","url_pdf":"http://arxiv.org/pdf/1711.02255v2.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":"convolutional-normalizing-flows","repo_url":"https://github.com/apsyx/mvae","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"variational-inference","task_name":"Variational Inference"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}