{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,12]],"date-time":"2025-11-12T14:04:44Z","timestamp":1762956284057,"version":"3.41.0"},"reference-count":44,"publisher":"Association for Computing Machinery (ACM)","issue":"1","license":[{"start":{"date-parts":[[2020,1,31]],"date-time":"2020-01-31T00:00:00Z","timestamp":1580428800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["Grant OAC-1910213 and Grant IIS-1845922"],"award-info":[{"award-number":["Grant OAC-1910213 and Grant IIS-1845922"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000183","name":"U.S. Army Research Office","doi-asserted-by":"crossref","award":["Grant W911NF-17-1-0485 and Grant W911NF-19-1-0162"],"award-info":[{"award-number":["Grant W911NF-17-1-0485 and Grant W911NF-19-1-0162"]}],"id":[{"id":"10.13039\/100000183","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Embed. Comput. Syst."],"published-print":{"date-parts":[[2020,1,31]]},"abstract":"<jats:p>Many real-world edge applications including object detection, robotics, and smart health are enabled by deploying deep neural networks (DNNs) on energy-constrained mobile platforms. In this article, we propose a novel approach to trade off energy and accuracy of inference at runtime using a design space called Learning Energy Accuracy Tradeoff Networks (LEANets). The key idea behind LEANets is to design classifiers of increasing complexity using pretrained DNNs to perform input-specific adaptive inference. The accuracy and energy consumption of the adaptive inference scheme depends on a set of thresholds, one for each classifier. To determine the set of threshold vectors to achieve different energy and accuracy tradeoffs, we propose a novel multiobjective optimization approach. We can select the appropriate threshold vector at runtime based on the desired tradeoff. We perform experiments on multiple pretrained DNNs including ConvNet, VGG-16, and MobileNet using diverse image classification datasets. Our results show that we get up to a 50% gain in energy for negligible loss in accuracy, and optimized LEANets achieve significantly better energy and accuracy tradeoff when compared to a state-of-the-art method referred to as Slimmable neural networks.<\/jats:p>","DOI":"10.1145\/3366636","type":"journal-article","created":{"date-parts":[[2020,2,7]],"date-time":"2020-02-07T07:03:59Z","timestamp":1581059039000},"page":"1-24","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":25,"title":["Design and Optimization of Energy-Accuracy Tradeoff Networks for Mobile Platforms via Pretrained Deep Models"],"prefix":"10.1145","volume":"19","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9715-393X","authenticated-orcid":false,"given":"Nitthilan Kanappan","family":"Jayakodi","sequence":"first","affiliation":[{"name":"School of Electrical Engineering and Computer Science, Washington State University, Pullman, WA"}]},{"given":"Syrine","family":"Belakaria","sequence":"additional","affiliation":[{"name":"School of Electrical Engineering and Computer Science, Washington State University, Pullman, WA"}]},{"given":"Aryan","family":"Deshwal","sequence":"additional","affiliation":[{"name":"School of Electrical Engineering and Computer Science, Washington State University, Pullman, WA"}]},{"given":"Janardhan Rao","family":"Doppa","sequence":"additional","affiliation":[{"name":"School of Electrical Engineering and Computer Science, Washington State University, Pullman, WA"}]}],"member":"320","published-online":{"date-parts":[[2020,2,6]]},"reference":[{"key":"e_1_2_1_1_1","unstructured":"Syrine Belakaria Aryan Deshwal and Janardhan Rao Doppa. 2019. Max-value entropy search for multi-objective Bayesian optimization. In Advances in Neural Information Processing Systems (NeurIPS\u201919).  Syrine Belakaria Aryan Deshwal and Janardhan Rao Doppa. 2019. Max-value entropy search for multi-objective Bayesian optimization. In Advances in Neural Information Processing Systems (NeurIPS\u201919)."},{"key":"e_1_2_1_2_1","unstructured":"Caffe-HRT. [n.d.]. Retrieved from https:\/\/github.com\/OAID\/Caffe-HRT.  Caffe-HRT. [n.d.]. Retrieved from https:\/\/github.com\/OAID\/Caffe-HRT."},{"key":"e_1_2_1_3_1","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO.2014.58"},{"key":"e_1_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2017.54"},{"key":"e_1_2_1_5_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.195"},{"volume-title":"BinaryNet: Training deep neural networks with weights and activations constrained to +1 or &minus;1. CoRR abs\/1602.02830","year":"2016","author":"Courbariaux Matthieu","key":"e_1_2_1_6_1"},{"key":"e_1_2_1_7_1","doi-asserted-by":"publisher","DOI":"10.1109\/4235.996017"},{"key":"e_1_2_1_8_1","doi-asserted-by":"publisher","DOI":"10.1145\/3358206"},{"key":"e_1_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.1109\/ASPDAC.2018.8297274"},{"key":"e_1_2_1_10_1","doi-asserted-by":"publisher","DOI":"10.5555\/2693068.2693078"},{"key":"e_1_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.5555\/2627435.2638577"},{"key":"e_1_2_1_12_1","doi-asserted-by":"publisher","DOI":"10.1145\/3093337.3037702"},{"volume-title":"Dally","year":"2016","author":"Han Song","key":"e_1_2_1_13_1"},{"volume-title":"Dally","year":"2015","author":"Han Song","key":"e_1_2_1_14_1"},{"volume-title":"Deep residual learning for image recognition. CoRR abs\/1512.03385","year":"2015","author":"He Kaiming","key":"e_1_2_1_15_1"},{"volume-title":"Amar Shah, and Ryan P. Adams.","year":"2016","author":"Hernandez-Lobato Daniel","key":"e_1_2_1_16_1"},{"volume-title":"NIPS Deep Learning Workshop.","year":"2014","author":"Hinton Geoffrey","key":"e_1_2_1_17_1"},{"volume-title":"MobileNets: Efficient convolutional neural networks for mobile vision applications. CoRR abs\/1704.04861","year":"2017","author":"Howard Andrew G.","key":"e_1_2_1_18_1"},{"volume-title":"Weinberger","year":"2016","author":"Huang Gao","key":"e_1_2_1_19_1"},{"volume-title":"SqueezeNet: AlexNet-level accuracy with 50\u00d7 fewer parameters and &lt;1MB model size. CoRR abs\/1602.07360","year":"2016","author":"Iandola Forrest N.","key":"e_1_2_1_20_1"},{"key":"e_1_2_1_21_1","first-page":"2881","article-title":"Trading-off accuracy and energy of deep inference on embedded systems: A co-design approach","volume":"37","author":"Jayakodi Nitthilan Kannappan","year":"2018","journal-title":"IEEE TCAD"},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1145\/2647868.2654889"},{"key":"e_1_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2018.2889053"},{"volume-title":"Proceedings of ISCA. 1--13","year":"2016","author":"Judd Patrick","key":"e_1_2_1_24_1"},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.1109\/TVLSI.2017.2700726"},{"volume-title":"IEEE Conference on Computer Vision and Pattern Recognition (CVPR\u201915)","author":"Lam Michael","key":"e_1_2_1_26_1"},{"key":"e_1_2_1_27_1","unstructured":"Haoxiang Li Zhe Lin Xiaohui Shen Jonathan Brandt and Gang Hua. 2015. A convolutional neural network cascade for face detection. In Proceeding of CVPR. 5325--5334.  Haoxiang Li Zhe Lin Xiaohui Shen Jonathan Brandt and Gang Hua. 2015. A convolutional neural network cascade for face detection. In Proceeding of CVPR. 5325--5334."},{"volume-title":"Proceedings of FPL. 1--8.","year":"2016","author":"Ma Yufei","key":"e_1_2_1_28_1"},{"key":"e_1_2_1_29_1","doi-asserted-by":"publisher","DOI":"10.1109\/TVLSI.2019.2926106"},{"key":"e_1_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.23919\/DATE.2018.8342068"},{"key":"e_1_2_1_31_1","unstructured":"ODROIOD-XU4. 2017. Retrieved March 29 2018 from https:\/\/wiki.odroid.com\/odroid-xu4\/hardware\/hardware.  ODROIOD-XU4. 2017. Retrieved March 29 2018 from https:\/\/wiki.odroid.com\/odroid-xu4\/hardware\/hardware."},{"key":"e_1_2_1_32_1","doi-asserted-by":"publisher","DOI":"10.3850\/9783981537079_0819"},{"volume-title":"ImageNet large scale visual recognition challenge. CoRR abs\/1409.0575","year":"2014","author":"Russakovsky Olga","key":"e_1_2_1_33_1"},{"key":"e_1_2_1_34_1","unstructured":"SmartPower2. [n.d.]. Retrieved from https:\/\/wiki.odroid.com\/accessory\/power_supply%_battery\/smartpower2.  SmartPower2. [n.d.]. Retrieved from https:\/\/wiki.odroid.com\/accessory\/power_supply%_battery\/smartpower2."},{"key":"e_1_2_1_35_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2016.2604288"},{"volume-title":"Seeger","year":"2010","author":"Srinivas Niranjan","key":"e_1_2_1_36_1"},{"volume-title":"Haocheng Fang, Sribhuvan Sajja, Mitchell Bognar, and Diana Marculescu.","year":"2018","author":"Stamoulis Dimitrios","key":"e_1_2_1_37_1"},{"key":"e_1_2_1_38_1","doi-asserted-by":"publisher","DOI":"10.1109\/JPROC.2017.2761740"},{"key":"e_1_2_1_39_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298594"},{"volume-title":"Cost-effective active learning for deep image classification. CoRR abs\/1701.03551","year":"2017","author":"Wang Keze","key":"e_1_2_1_40_1"},{"key":"e_1_2_1_41_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.643"},{"volume-title":"Huang","year":"2019","author":"Yu Jiahui","key":"e_1_2_1_42_1"},{"volume-title":"Incremental network quantization: Towards lossless CNNs with low-precision weights. Arxiv:1702.03044","year":"2017","author":"Zhou Aojun","key":"e_1_2_1_43_1"},{"volume-title":"Evolutionary Algorithms for Multiobjective Optimization: Methods and Applications","author":"Zitzler Eckart","key":"e_1_2_1_44_1"}],"container-title":["ACM Transactions on Embedded Computing Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3366636","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3366636","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3366636","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T23:13:34Z","timestamp":1750202014000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3366636"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,1,31]]},"references-count":44,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2020,1,31]]}},"alternative-id":["10.1145\/3366636"],"URL":"https:\/\/doi.org\/10.1145\/3366636","relation":{},"ISSN":["1539-9087","1558-3465"],"issn-type":[{"type":"print","value":"1539-9087"},{"type":"electronic","value":"1558-3465"}],"subject":[],"published":{"date-parts":[[2020,1,31]]},"assertion":[{"value":"2019-08-01","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2019-09-01","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2020-02-06","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}