{"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/sifting-common-information-from-many","title":"Sifting Common Information from Many Variables","arxiv_id":"1606.02307","date":"2016-06-07","proceeding":null,"authors":["Greg Ver Steeg","Shuyang Gao","Kyle Reing","Aram Galstyan"],"abstract":"Measuring the relationship between any pair of variables is a rich and active\narea of research that is central to scientific practice. In contrast,\ncharacterizing the common information among any group of variables is typically\na theoretical exercise with few practical methods for high-dimensional data. A\npromising solution would be a multivariate generalization of the famous Wyner\ncommon information, but this approach relies on solving an apparently\nintractable optimization problem. We leverage the recently introduced\ninformation sieve decomposition to formulate an incremental version of the\ncommon information problem that admits a simple fixed point solution, fast\nconvergence, and complexity that is linear in the number of variables. This\nscalable approach allows us to demonstrate the usefulness of common information\nin high-dimensional learning problems. The sieve outperforms standard methods\non dimensionality reduction tasks, solves a blind source separation problem\nthat cannot be solved with ICA, and accurately recovers structure in brain\nimaging data.","url_abs":"http://arxiv.org/abs/1606.02307v4","url_pdf":"http://arxiv.org/pdf/1606.02307v4.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":"sifting-common-information-from-many","repo_url":"https://github.com/gregversteeg/LinearSieve","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"dimensionality-reduction","task_name":"Dimensionality Reduction"},{"task_slug":"blind-source-separation","task_name":"blind source separation"}],"methods":[{"method_slug":"ica","method_name":"ICA"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}