{"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/efficient-use-of-limited-memory-accelerators","title":"Efficient Use of Limited-Memory Accelerators for Linear Learning on Heterogeneous Systems","arxiv_id":"1708.05357","date":"2017-08-17","proceeding":"NeurIPS 2017 12","authors":["Celestine Dünner","Thomas Parnell","Martin Jaggi"],"abstract":"We propose a generic algorithmic building block to accelerate training of\nmachine learning models on heterogeneous compute systems. Our scheme allows to\nefficiently employ compute accelerators such as GPUs and FPGAs for the training\nof large-scale machine learning models, when the training data exceeds their\nmemory capacity. Also, it provides adaptivity to any system's memory hierarchy\nin terms of size and processing speed. Our technique is built upon novel\ntheoretical insights regarding primal-dual coordinate methods, and uses duality\ngap information to dynamically decide which part of the data should be made\navailable for fast processing. To illustrate the power of our approach we\ndemonstrate its performance for training of generalized linear models on a\nlarge-scale dataset exceeding the memory size of a modern GPU, showing an\norder-of-magnitude speedup over existing approaches.","url_abs":"http://arxiv.org/abs/1708.05357v2","url_pdf":"http://arxiv.org/pdf/1708.05357v2.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":"efficient-use-of-limited-memory-accelerators","repo_url":"https://github.com/ElizaWszola/HTHC","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"machine-learning","task_name":"BIG-bench Machine Learning"},{"task_slug":null,"task_name":"GPU"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}