{"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/unbiased-implicit-variational-inference","title":"Unbiased Implicit Variational Inference","arxiv_id":"1808.02078","date":"2018-08-06","proceeding":null,"authors":["Michalis K. Titsias","Francisco J. R. Ruiz"],"abstract":"We develop unbiased implicit variational inference (UIVI), a method that\nexpands the applicability of variational inference by defining an expressive\nvariational family. UIVI considers an implicit variational distribution\nobtained in a hierarchical manner using a simple reparameterizable distribution\nwhose variational parameters are defined by arbitrarily flexible deep neural\nnetworks. Unlike previous works, UIVI directly optimizes the evidence lower\nbound (ELBO) rather than an approximation to the ELBO. We demonstrate UIVI on\nseveral models, including Bayesian multinomial logistic regression and\nvariational autoencoders, and show that UIVI achieves both tighter ELBO and\nbetter predictive performance than existing approaches at a similar\ncomputational cost.","url_abs":"http://arxiv.org/abs/1808.02078v3","url_pdf":"http://arxiv.org/pdf/1808.02078v3.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":"unbiased-implicit-variational-inference","repo_url":"https://github.com/franrruiz/uivi","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"variational-inference","task_name":"Variational Inference"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[{"method_slug":"logistic-regression","method_name":"Logistic Regression"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1808.02078","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}