{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,25]],"date-time":"2026-10-25T17:42:57Z","timestamp":1792950177364,"version":"4.5.0"},"reference-count":51,"publisher":"Association for Computing Machinery (ACM)","issue":"4","license":[{"start":{"date-parts":[[2019,10,3]],"date-time":"2019-10-03T00:00:00Z","timestamp":1570060800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["Grant 61874124, Grant 61876173, Grant 61432017, Grant 61532017"],"award-info":[{"award-number":["Grant 61874124, Grant 61876173, Grant 61432017, Grant 61532017"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"name":"YESS hip program","award":["No.YESS2016qnrc001"],"award-info":[{"award-number":["No.YESS2016qnrc001"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["J. Emerg. Technol. Comput. Syst."],"published-print":{"date-parts":[[2019,10,31]]},"abstract":"<jats:p>Recently the development of deep learning has been propelling the sheer growth of vision and speech applications on lightweight embedded and mobile systems. However, the limitation of computation resource and power delivery capability in embedded platforms is recognized as a significant bottleneck that prevents the systems from providing real-time deep learning ability, since the inference of deep convolutional neural networks (CNNs) and recurrent neural networks (RNNs) involves large quantities of weights and operations. Particularly, how to provide quality-of-services (QoS)-guaranteed neural network inference ability in the multitask execution environment of multicore SoCs is even more complicated due to the existence of resource contention. In this article, we present a novel deep neural network architecture, MV-Net, which provides performance elasticity and contention-aware self-scheduling ability for QoS enhancement in mobile computing systems. When the constraints of QoS, output accuracy, and resource contention status of the system change, MV-Net can dynamically reconfigure the corresponding neural network propagation paths and thus achieves an effective tradeoff between neural network computational complexity and prediction accuracy via approximate computing. The experimental results show that (1) MV-Net significantly improves the performance flexibility of current CNN models and makes it possible to provide always-guaranteed QoS in a multitask environment, and (2) it satisfies the quality-of-results (QoR) requirement, outperforming the baseline implementation significantly, and improves the system energy efficiency at the same time.<\/jats:p>","DOI":"10.1145\/3358696","type":"journal-article","created":{"date-parts":[[2019,10,3]],"date-time":"2019-10-03T12:18:58Z","timestamp":1570105138000},"page":"1-25","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":4,"title":["MV-Net"],"prefix":"10.1145","volume":"15","author":[{"given":"Yibin","family":"Tang","sequence":"first","affiliation":[{"name":"State Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences; University of Chinese Academy of Sciences"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ying","family":"Wang","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences; University of Chinese Academy of Sciences"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Huawei","family":"Li","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences; University of Chinese Academy of Sciences; Peng Cheng Laboratory"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaowei","family":"Li","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences; University of Chinese Academy of Sciences"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2019,10,3]]},"reference":[{"key":"e_1_2_1_1_1","first-page":"06834","article-title":"Skip RNN: Learning to skip state updates in recurrent neural networks","volume":"1708","author":"Campos V\u00edctor","year":"2017","journal-title":"Arxiv Preprint Arxiv"},{"key":"e_1_2_1_2_1","doi-asserted-by":"publisher","DOI":"10.1109\/IISWC.2009.5306797"},{"key":"e_1_2_1_3_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2014.414"},{"key":"e_1_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.327"},{"key":"e_1_2_1_5_1","unstructured":"Arm Cortex. A57 MPCore processor technical reference manual infocenter. arm. com arithmetic. Logical Unit Advanced SIMD Micro-Operation Vector Floating Point.  Arm Cortex. A57 MPCore processor technical reference manual infocenter. arm. com arithmetic. Logical Unit Advanced SIMD Micro-Operation Vector Floating Point."},{"key":"e_1_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.316"},{"key":"e_1_2_1_7_1","volume-title":"2014 19th Asia and South Pacific Design Automation Conference (ASP-DAC\u201914)","author":"Du Zidong","year":"2014"},{"key":"e_1_2_1_8_1","volume-title":"2013 IEEE 34th Real-Time Systems Symposium (RTSS'13)","author":"Elliott Glenn A."},{"key":"e_1_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2010.5539906"},{"key":"e_1_2_1_10_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.future.2013.01.010"},{"key":"e_1_2_1_11_1","first-page":"00149","article-title":"Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding","volume":"1510","author":"Han Song","year":"2015","journal-title":"Arxiv Preprint Arxiv"},{"key":"e_1_2_1_12_1","unstructured":"Song Han Jeff Pool John Tran and William Dally. 2015. Learning both weights and connections for efficient neural network. In Advances in Neural Information Processing Systems (NIPS\u201915). 1135--1143.  Song Han Jeff Pool John Tran and William Dally. 2015. Learning both weights and connections for efficient neural network. In Advances in Neural Information Processing Systems (NIPS\u201915). 1135--1143."},{"key":"e_1_2_1_13_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7299173"},{"key":"e_1_2_1_14_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.1145\/2228360.2228513"},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.1145\/2647868.2654889"},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1109\/RTSS.2011.13"},{"key":"e_1_2_1_18_1","volume-title":"Proc. USENIX Annual Technical Conference (USENIX ATC\u201911)","author":"Kato Shinpei","year":"2011"},{"key":"e_1_2_1_19_1","volume-title":"Hinton","author":"Krizhevsky Alex","year":"2012"},{"key":"e_1_2_1_20_1","doi-asserted-by":"crossref","unstructured":"Siwei Lai Liheng Xu Kang Liu and Jun Zhao. 2015. Recurrent convolutional neural networks for text classification. In AAAI. 2267--2273.  Siwei Lai Liheng Xu Kang Liu and Jun Zhao. 2015. Recurrent convolutional neural networks for text classification. In AAAI. 2267--2273.","DOI":"10.1609\/aaai.v29i1.9513"},{"key":"e_1_2_1_21_1","volume-title":"Proceedings of the 16th International Workshop on Mobile Computing Systems and Applications (IWMCSA\u201915)","author":"Nicholas"},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2016.2547916"},{"key":"e_1_2_1_23_1","unstructured":"Ji Lin Yongming Rao Jiwen Lu and Jie Zhou. 2017. Runtime neural pruning. In Advances in Neural Information Processing Systems (NIPS\u201917). 2181--2191.  Ji Lin Yongming Rao Jiwen Lu and Jie Zhou. 2017. Runtime neural pruning. In Advances in Neural Information Processing Systems (NIPS\u201917). 2181--2191."},{"key":"e_1_2_1_24_1","first-page":"4400","article-title":"Network in network","volume":"1312","author":"Lin Min","year":"2013","journal-title":"Arxiv Preprint Arxiv"},{"key":"e_1_2_1_25_1","first-page":"02325","article-title":"Ssd: Single shot multibox detector","volume":"1512","author":"Liu Wei","year":"2015","journal-title":"ArXiv Preprint Arxiv"},{"key":"e_1_2_1_26_1","first-page":"06217","article-title":"Deciding how to decide: Dynamic routing in artificial neural networks","volume":"1703","author":"Mcgill Mason","year":"2017","journal-title":"ArXiv Preprint ArXiv"},{"key":"e_1_2_1_27_1","unstructured":"Guido F. Montufar Razvan Pascanu Kyunghyun Cho and Yoshua Bengio. 2014. On the number of linear regions of deep neural networks. In Advances in Neural Information Processing Systems (NIPS\u201914). 2924--2932.  Guido F. Montufar Razvan Pascanu Kyunghyun Cho and Yoshua Bengio. 2014. On the number of linear regions of deep neural networks. In Advances in Neural Information Processing Systems (NIPS\u201914). 2924--2932."},{"key":"e_1_2_1_28_1","unstructured":"Nvidia. 2015. Jetson tx1 module. http:\/\/www.nvidia.com\/object\/embedded-systems-dev-kits-modules.html.  Nvidia. 2015. Jetson tx1 module. http:\/\/www.nvidia.com\/object\/embedded-systems-dev-kits-modules.html."},{"key":"e_1_2_1_29_1","volume-title":"Chung","author":"Ovtcharov Kalin","year":"2015"},{"key":"e_1_2_1_30_1","volume-title":"FALCON: Feature driven selective classification for energy-efficient image recognition","author":"Panda Priyadarshini","year":"2016"},{"key":"e_1_2_1_31_1","unstructured":"Priyadarshini Panda Abhronil Sengupta and Kaushik Roy. 2015. Conditional deep learning for energy-efficient and enhanced pattern recognition. In Design Automation 8 Test in Europe Conference 8 Exhibition (DATE\u201915). 36.  Priyadarshini Panda Abhronil Sengupta and Kaushik Roy. 2015. Conditional deep learning for energy-efficient and enhanced pattern recognition. In Design Automation 8 Test in Europe Conference 8 Exhibition (DATE\u201915). 36."},{"key":"e_1_2_1_32_1","doi-asserted-by":"publisher","DOI":"10.1145\/3007192"},{"key":"e_1_2_1_33_1","doi-asserted-by":"publisher","DOI":"10.1145\/2847263.2847265"},{"key":"e_1_2_1_34_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.91"},{"key":"e_1_2_1_35_1","volume-title":"Memory system architecture for real-time multitasking systems","author":"Rixner Scott"},{"key":"e_1_2_1_36_1","doi-asserted-by":"publisher","DOI":"10.1109\/ISBA.2016.7477230"},{"key":"e_1_2_1_37_1","first-page":"04582","article-title":"Query-reduction networks for question answering","volume":"1606","author":"Seo Minjoon","year":"2016","journal-title":"ArXiv Preprint Arxiv"},{"key":"e_1_2_1_38_1","doi-asserted-by":"publisher","DOI":"10.1145\/2897937.2897995"},{"key":"e_1_2_1_39_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"e_1_2_1_40_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2014.6854363"},{"key":"e_1_2_1_41_1","doi-asserted-by":"publisher","DOI":"10.1145\/2744769.2744904"},{"key":"e_1_2_1_42_1","doi-asserted-by":"publisher","DOI":"10.1145\/2367589.2367596"},{"key":"e_1_2_1_43_1","doi-asserted-by":"publisher","DOI":"10.1109\/ASPDAC.2017.7858304"},{"key":"e_1_2_1_44_1","first-page":"06994","article-title":"Structured probabilistic pruning for convolutional neural network acceleration","volume":"1709","author":"Wang Huan","year":"2017","journal-title":"ArXiv Preprint Arxiv"},{"key":"e_1_2_1_45_1","volume-title":"Proceedings of the IEEE Design Automation Conference (DAC\u201917)","author":"Wang Ying","year":"2017"},{"key":"e_1_2_1_46_1","doi-asserted-by":"publisher","DOI":"10.1145\/2897937.2898003"},{"key":"e_1_2_1_47_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00919"},{"key":"e_1_2_1_48_1","volume-title":"Proton Testing of nVidia Jetson TX1. Retrieved on","author":"Wyrwas Edward J.","year":"2016"},{"key":"e_1_2_1_49_1","doi-asserted-by":"publisher","DOI":"10.1145\/2925426.2926265"},{"key":"e_1_2_1_50_1","doi-asserted-by":"publisher","DOI":"10.1109\/LCA.2014.2319079"},{"key":"e_1_2_1_51_1","doi-asserted-by":"publisher","DOI":"10.1109\/RTAS.2015.7108420"}],"container-title":["ACM Journal on Emerging Technologies in Computing Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3358696","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3358696","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T23:13:26Z","timestamp":1750202006000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3358696"}},"subtitle":["Toward Real-Time Deep Learning on Mobile GPGPU Systems"],"short-title":[],"issued":{"date-parts":[[2019,10,3]]},"references-count":51,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2019,10,31]]}},"alternative-id":["10.1145\/3358696"],"URL":"https:\/\/doi.org\/10.1145\/3358696","relation":{},"ISSN":["1550-4832","1550-4840"],"issn-type":[{"value":"1550-4832","type":"print"},{"value":"1550-4840","type":"electronic"}],"subject":[],"published":{"date-parts":[[2019,10,3]]},"assertion":[{"value":"2018-07-01","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2019-07-01","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2019-10-03","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}