{"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/understanding-top-k-sparsification-in-1","title":"Understanding Top-k Sparsification in Distributed Deep Learning","arxiv_id":"1911.08772","date":"2019-11-20","proceeding":null,"authors":["Shaohuai Shi","Xiaowen Chu","Ka Chun Cheung","Simon See"],"abstract":"Distributed stochastic gradient descent (SGD) algorithms are widely deployed in training large-scale deep learning models, while the communication overhead among workers becomes the new system bottleneck. Recently proposed gradient sparsification techniques, especially Top-$k$ sparsification with error compensation (TopK-SGD), can significantly reduce the communication traffic without an obvious impact on the model accuracy. Some theoretical studies have been carried out to analyze the convergence property of TopK-SGD. However, existing studies do not dive into the details of Top-$k$ operator in gradient sparsification and use relaxed bounds (e.g., exact bound of Random-$k$) for analysis; hence the derived results cannot well describe the real convergence performance of TopK-SGD. To this end, we first study the gradient distributions of TopK-SGD during the training process through extensive experiments. We then theoretically derive a tighter bound for the Top-$k$ operator. Finally, we exploit the property of gradient distribution to propose an approximate top-$k$ selection algorithm, which is computing-efficient for GPUs, to improve the scaling efficiency of TopK-SGD by significantly reducing the computing overhead. Codes are available at: \\url{https://github.com/hclhkbu/GaussianK-SGD}.","url_abs":"https://arxiv.org/abs/1911.08772v1","url_pdf":"https://arxiv.org/pdf/1911.08772v1.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":"understanding-top-k-sparsification-in-1","repo_url":"https://github.com/hclhkbu/GaussianK-SGD","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"deep-learning","task_name":"Deep Learning"}],"methods":[{"method_slug":"gradient-sparsification","method_name":"Gradient Sparsification"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1911.08772","atlas_url":"https://app.syntology.ai/?focus=1911.08772","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}