{"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/axnet-approximate-computing-using-an-end-to","title":"AXNet: ApproXimate computing using an end-to-end trainable neural network","arxiv_id":"1807.10458","date":"2018-07-27","proceeding":null,"authors":["Zhenghao Peng","Xuyang Chen","Chengwen Xu","Naifeng Jing","Xiaoyao Liang","Cewu Lu","Li Jiang"],"abstract":"Neural network based approximate computing is a universal architecture\npromising to gain tremendous energy-efficiency for many error resilient\napplications. To guarantee the approximation quality, existing works deploy two\nneural networks (NNs), e.g., an approximator and a predictor. The approximator\nprovides the approximate results, while the predictor predicts whether the\ninput data is safe to approximate with the given quality requirement. However,\nit is non-trivial and time-consuming to make these two neural network\ncoordinate---they have different optimization objectives---by training them\nseparately. This paper proposes a novel neural network structure---AXNet---to\nfuse two NNs to a holistic end-to-end trainable NN. Leveraging the philosophy\nof multi-task learning, AXNet can tremendously improve the invocation\n(proportion of safe-to-approximate samples) and reduce the approximation error.\nThe training effort also decrease significantly. Experiment results show 50.7%\nmore invocation and substantial cuts of training time when compared to existing\nneural network based approximate computing framework.","url_abs":"http://arxiv.org/abs/1807.10458v2","url_pdf":"http://arxiv.org/pdf/1807.10458v2.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":"axnet-approximate-computing-using-an-end-to","repo_url":"https://github.com/ACA-Lab-SJTU/approximate-computing","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"axnet-approximate-computing-using-an-end-to","repo_url":"https://github.com/PengZhenghao/AXNet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"},{"task_slug":"philosophy","task_name":"Philosophy"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}