{"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/detecting-dependencies-in-sparse-multivariate","title":"Detecting Dependencies in Sparse, Multivariate Databases Using Probabilistic Programming and Non-parametric Bayes","arxiv_id":"1611.01708","date":"2016-11-05","proceeding":null,"authors":["Feras Saad","Vikash Mansinghka"],"abstract":"Datasets with hundreds of variables and many missing values are commonplace.\nIn this setting, it is both statistically and computationally challenging to\ndetect true predictive relationships between variables and also to suppress\nfalse positives. This paper proposes an approach that combines probabilistic\nprogramming, information theory, and non-parametric Bayes. It shows how to use\nBayesian non-parametric modeling to (i) build an ensemble of joint probability\nmodels for all the variables; (ii) efficiently detect marginal independencies;\nand (iii) estimate the conditional mutual information between arbitrary subsets\nof variables, subject to a broad class of constraints. Users can access these\ncapabilities using BayesDB, a probabilistic programming platform for\nprobabilistic data analysis, by writing queries in a simple, SQL-like language.\nThis paper demonstrates empirically that the method can (i) detect\ncontext-specific (in)dependencies on challenging synthetic problems and (ii)\nyield improved sensitivity and specificity over baselines from statistics and\nmachine learning, on a real-world database of over 300 sparsely observed\nindicators of macroeconomic development and public health.","url_abs":"http://arxiv.org/abs/1611.01708v2","url_pdf":"http://arxiv.org/pdf/1611.01708v2.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":"detecting-dependencies-in-sparse-multivariate","repo_url":"https://github.com/probcomp/bayeslite","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"missing-values","task_name":"Missing Values"},{"task_slug":"probabilistic-programming","task_name":"Probabilistic Programming"},{"task_slug":"specificity","task_name":"Specificity"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}