{"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/lirs-enabling-efficient-machine-learning-on","title":"LIRS: Enabling efficient machine learning on NVM-based storage via a lightweight implementation of random shuffling","arxiv_id":"1810.04509","date":"2018-10-10","proceeding":null,"authors":["Zhi-Lin Ke","Hsiang-Yun Cheng","Chia-Lin Yang"],"abstract":"Machine learning algorithms, such as Support Vector Machine (SVM) and Deep\nNeural Network (DNN), have gained a lot of interests recently. When training a\nmachine learning algorithm, randomly shuffle all the training data can improve\nthe testing accuracy and boost the convergence rate. Nevertheless, realizing\ntraining data random shuffling in a real system is not a straightforward\nprocess due to the slow random accesses in hard disk drive (HDD). To avoid\nfrequent random disk access, the effect of random shuffling is often limited in\nexisting approaches. With the emerging non-volatile memory-based storage\ndevice, such as Intel Optane SSD, which provides fast random accesses, we\npropose a lightweight implementation of random shuffling (LIRS) to randomly\nshuffle the indexes of the entire training dataset, and the selected training\ninstances are directly accessed from the storage and packed into batches.\nExperimental results show that LIRS can reduce the total training time of SVM\nand DNN by 49.9% and 43.5% on average, and improve the final testing accuracy\non DNN by 1.01%.","url_abs":"http://arxiv.org/abs/1810.04509v1","url_pdf":"http://arxiv.org/pdf/1810.04509v1.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":"lirs-enabling-efficient-machine-learning-on","repo_url":"https://github.com/winiel559/ZhiLin-LIRS","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"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"non-maximum-suppression","method_name":"Non Maximum Suppression"},{"method_slug":"ssd","method_name":"SSD"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}