{"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/slimnets-an-exploration-of-deep-model","title":"SlimNets: An Exploration of Deep Model Compression and Acceleration","arxiv_id":"1808.00496","date":"2018-08-01","proceeding":null,"authors":["Ini Oguntola","Subby Olubeko","Christopher Sweeney"],"abstract":"Deep neural networks have achieved increasingly accurate results on a wide\nvariety of complex tasks. However, much of this improvement is due to the\ngrowing use and availability of computational resources (e.g use of GPUs, more\nlayers, more parameters, etc). Most state-of-the-art deep networks, despite\nperforming well, over-parameterize approximate functions and take a significant\namount of time to train. With increased focus on deploying deep neural networks\non resource constrained devices like smart phones, there has been a push to\nevaluate why these models are so resource hungry and how they can be made more\nefficient. This work evaluates and compares three distinct methods for deep\nmodel compression and acceleration: weight pruning, low rank factorization, and\nknowledge distillation. Comparisons on VGG nets trained on CIFAR10 show that\neach of the models on their own are effective, but that the true power lies in\ncombining them. We show that by combining pruning and knowledge distillation\nmethods we can create a compressed network 85 times smaller than the original,\nall while retaining 96% of the original model's accuracy.","url_abs":"http://arxiv.org/abs/1808.00496v1","url_pdf":"http://arxiv.org/pdf/1808.00496v1.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":"slimnets-an-exploration-of-deep-model","repo_url":"https://github.com/ChristopherSweeney/SlimNets","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"knowledge-distillation","task_name":"Knowledge Distillation"},{"task_slug":"model-compression","task_name":"Model Compression"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"pruning","method_name":"Pruning"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}