{"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/optimization-for-large-scale-machine-learning","title":"Optimization for Large-Scale Machine Learning with Distributed Features and Observations","arxiv_id":"1610.10060","date":"2016-10-31","proceeding":null,"authors":["Alexandros Nathan","Diego Klabjan"],"abstract":"As the size of modern data sets exceeds the disk and memory capacities of a\nsingle computer, machine learning practitioners have resorted to parallel and\ndistributed computing. Given that optimization is one of the pillars of machine\nlearning and predictive modeling, distributed optimization methods have\nrecently garnered ample attention in the literature. Although previous research\nhas mostly focused on settings where either the observations, or features of\nthe problem at hand are stored in distributed fashion, the situation where both\nare partitioned across the nodes of a computer cluster (doubly distributed) has\nbarely been studied. In this work we propose two doubly distributed\noptimization algorithms. The first one falls under the umbrella of distributed\ndual coordinate ascent methods, while the second one belongs to the class of\nstochastic gradient/coordinate descent hybrid methods. We conduct numerical\nexperiments in Spark using real-world and simulated data sets and study the\nscaling properties of our methods. Our empirical evaluation of the proposed\nalgorithms demonstrates the out-performance of a block distributed ADMM method,\nwhich, to the best of our knowledge is the only other existing doubly\ndistributed optimization algorithm.","url_abs":"http://arxiv.org/abs/1610.10060v2","url_pdf":"http://arxiv.org/pdf/1610.10060v2.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":"optimization-for-large-scale-machine-learning","repo_url":"https://github.com/anathan90/RADiSA","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"machine-learning","task_name":"BIG-bench Machine Learning"},{"task_slug":"distributed-computing","task_name":"Distributed Computing"},{"task_slug":"distributed-optimization","task_name":"Distributed Optimization"}],"methods":[{"method_slug":"admm","method_name":"ADMM"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}