{"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/filtering-variational-objectives","title":"Filtering Variational Objectives","arxiv_id":"1705.09279","date":"2017-05-25","proceeding":"NeurIPS 2017 12","authors":["Chris J. Maddison","Dieterich Lawson","George Tucker","Nicolas Heess","Mohammad Norouzi","andriy mnih","Arnaud Doucet","Yee Whye Teh"],"abstract":"When used as a surrogate objective for maximum likelihood estimation in\nlatent variable models, the evidence lower bound (ELBO) produces\nstate-of-the-art results. Inspired by this, we consider the extension of the\nELBO to a family of lower bounds defined by a particle filter's estimator of\nthe marginal likelihood, the filtering variational objectives (FIVOs). FIVOs\ntake the same arguments as the ELBO, but can exploit a model's sequential\nstructure to form tighter bounds. We present results that relate the tightness\nof FIVO's bound to the variance of the particle filter's estimator by\nconsidering the generic case of bounds defined as log-transformed likelihood\nestimators. Experimentally, we show that training with FIVO results in\nsubstantial improvements over training the same model architecture with the\nELBO on sequential data.","url_abs":"http://arxiv.org/abs/1705.09279v3","url_pdf":"http://arxiv.org/pdf/1705.09279v3.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":"filtering-variational-objectives","repo_url":"https://github.com/AdrienCorenflos/differentiableFIVO","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"filtering-variational-objectives","repo_url":"https://github.com/tensorflow/models","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"filtering-variational-objectives","repo_url":"https://github.com/tensorflow/models/tree/master/research/fivo","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1705.09279","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}