{"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/invocation-driven-neural-approximate","title":"Invocation-driven Neural Approximate Computing with a Multiclass-Classifier and Multiple Approximators","arxiv_id":"1810.08379","date":"2018-10-19","proceeding":null,"authors":["Haiyue Song","Chengwen Xu","Qiang Xu","Zhuoran Song","Naifeng Jing","Xiaoyao Liang","Li Jiang"],"abstract":"Neural approximate computing gains enormous energy-efficiency at the cost of\ntolerable quality-loss. A neural approximator can map the input data to output\nwhile a classifier determines whether the input data are safe to approximate\nwith quality guarantee. However, existing works cannot maximize the invocation\nof the approximator, resulting in limited speedup and energy saving. By\nexploring the mapping space of those target functions, in this paper, we\nobserve a nonuniform distribution of the approximation error incurred by the\nsame approximator. We thus propose a novel approximate computing architecture\nwith a Multiclass-Classifier and Multiple Approximators (MCMA). These\napproximators have identical network topologies and thus can share the same\nhardware resource in a neural processing unit(NPU) clip. In the runtime, MCMA\ncan swap in the invoked approximator by merely shipping the synapse weights\nfrom the on-chip memory to the buffers near MAC within a cycle. We also propose\nefficient co-training methods for such MCMA architecture. Experimental results\nshow a more substantial invocation of MCMA as well as the gain of\nenergy-efficiency.","url_abs":"http://arxiv.org/abs/1810.08379v1","url_pdf":"http://arxiv.org/pdf/1810.08379v1.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":"invocation-driven-neural-approximate","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}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}