{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,7,30]],"date-time":"2025-07-30T13:40:47Z","timestamp":1753882847325,"version":"3.41.2"},"reference-count":25,"publisher":"World Scientific Pub Co Pte Ltd","issue":"15","funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62072067","61602080"],"award-info":[{"award-number":["62072067","61602080"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Dalian Young Star of Science and Technology Project","award":["2021RQ055"],"award-info":[{"award-number":["2021RQ055"]}]},{"name":"the Xinjiang Uygur Autonomous Region Natural Science Foundation Project","award":["2022D01A226"],"award-info":[{"award-number":["2022D01A226"]}]},{"DOI":"10.13039\/501100010245","name":"Social Science Foundation of Liaoning Province","doi-asserted-by":"publisher","award":["L17CTQ002"],"award-info":[{"award-number":["L17CTQ002"]}],"id":[{"id":"10.13039\/501100010245","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100012226","name":"Fundamental Research Funds for the Central Universities","doi-asserted-by":"publisher","award":["DUT21JC27"],"award-info":[{"award-number":["DUT21JC27"]}],"id":[{"id":"10.13039\/501100012226","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["U1808206"],"award-info":[{"award-number":["U1808206"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61872052"],"award-info":[{"award-number":["61872052"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61602084"],"award-info":[{"award-number":["61602084"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["J CIRCUIT SYST COMP"],"published-print":{"date-parts":[[2024,10]]},"abstract":"<jats:p> Embedded sensing devices that are battery-operated and resource-limited are increasingly utilizing recurrent neural networks (RNNs) to attain real-time perception of the complex environment. Nevertheless, the majority of RNNs typically handle substantial amounts of time-series data, resulting in computational intensity and high energy consumption. Hence, the development of energy-efficient RNN models is a crucial focus. This paper introduces IS-RNN, an energy-efficient RNN model featuring importance-based sparsification, aimed at enhancing the energy efficiency of RNN-based in-sensor inference. The principal motivation for this research arises from the recognition that time series data obtained from sensor sensing often exhibit inherent redundancy. Leveraging input sequence sparsification allows for the elimination of invalid input accesses and computations, thus optimizing efficiency. As a result, optimizing the sparsity of the input sequence through adaptive control allows us to improve inference performance while adhering to energy constraints. In particular, we build an importance score analysis method to quantify the contribution of each input toward the RNN inference performance. Building upon the proposed quantitative importance score analysis, we introduce a dynamic efficient RNN architecture, which enables the model to make decisions based on important data to obtain the highest accuracy under energy constraints. Moreover, we introduce an adaptive RNN inference method to enhance prediction accuracy under energy constraints by making a trade-off between computational cost and model performance. We evaluated IS-RNN on five datasets and the results show that it achieves a higher inference accuracy with the same amount of data processed compared to the state-of-the-art methods. <\/jats:p>","DOI":"10.1142\/s0218126624502712","type":"journal-article","created":{"date-parts":[[2024,4,18]],"date-time":"2024-04-18T09:13:50Z","timestamp":1713431630000},"source":"Crossref","is-referenced-by-count":0,"title":["Efficient Recurrent Neural Networks via Importance-Based Sparsification"],"prefix":"10.1142","volume":"33","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6289-1513","authenticated-orcid":false,"given":"Jiankang","family":"Ren","sequence":"first","affiliation":[{"name":"School of Computer Science and Technology, Dalian University of Technology, Dalian 116024, P. R. China"},{"name":"Key Laboratory of Social Computing and Cognitive Intelligence, (Dalian University of Technology), Ministry of Education, Dalian 116024, P. R. China"},{"name":"School of Computer Science and Technology, Xinjiang Normal University, Urumqi 830054, P. R. China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-5966-1159","authenticated-orcid":false,"given":"Zheng","family":"Ni","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Dalian University of Technology, Dalian 116024, P. R. China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4393-7461","authenticated-orcid":false,"given":"Xiaoyan","family":"Su","sequence":"additional","affiliation":[{"name":"School of Economics and Management, Dalian University of Technology, Dalian 116024, P. R. China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6823-7077","authenticated-orcid":false,"given":"Haijun","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Xinjiang Normal University, Urumqi 830054, P. R. China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-9924-9603","authenticated-orcid":false,"given":"Haifang","family":"Li","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Xinjiang Normal University, Urumqi 830054, P. R. China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-9768-1739","authenticated-orcid":false,"given":"Shengyu","family":"Li","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Dalian University of Technology, Dalian 116024, P. R. China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"219","published-online":{"date-parts":[[2024,6,10]]},"reference":[{"key":"S0218126624502712BIB001","doi-asserted-by":"publisher","DOI":"10.1109\/JSEN.2023.3286580"},{"key":"S0218126624502712BIB002","doi-asserted-by":"publisher","DOI":"10.1016\/j.measurement.2023.112505"},{"key":"S0218126624502712BIB003","doi-asserted-by":"publisher","DOI":"10.1109\/ICMLA58977.2023.00073"},{"key":"S0218126624502712BIB004","doi-asserted-by":"publisher","DOI":"10.1109\/TVT.2023.3292401"},{"key":"S0218126624502712BIB005","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2021.3129114"},{"key":"S0218126624502712BIB006","doi-asserted-by":"publisher","DOI":"10.1145\/3448974"},{"key":"S0218126624502712BIB007","doi-asserted-by":"publisher","DOI":"10.1109\/RTAS52030.2021.00020"},{"key":"S0218126624502712BIB008","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-020-09816-7"},{"key":"S0218126624502712BIB009","doi-asserted-by":"publisher","DOI":"10.1109\/SLT54892.2023.10022991"},{"key":"S0218126624502712BIB011","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2022.108797"},{"key":"S0218126624502712BIB012","first-page":"10976","volume-title":"The Thirty-Second Annual Conference on Neural Information Processing Systems (NIPS)","volume":"31","author":"Dennis D.","year":"2018"},{"key":"S0218126624502712BIB013","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746863"},{"key":"S0218126624502712BIB015","first-page":"1","volume-title":"The Thirtieth Annual Conference on Neural Information Processing Systems (NIPS)","author":"Neil D.","year":"2016"},{"key":"S0218126624502712BIB017","first-page":"1","volume-title":"The Ninth International Conference on Learning Representations (ICLR)","author":"Mayet T.","year":"2021"},{"key":"S0218126624502712BIB018","first-page":"1135","volume-title":"The Twenty-Eighth International Conference on Neural Information Processing Systems (NIPS)","volume":"28","author":"Han S.","year":"2015"},{"key":"S0218126624502712BIB019","first-page":"129","volume-title":"The Seventh Annual Conference on Machine Learning and Systems (MLSys)","volume":"2","author":"Blalock D.","year":"2020"},{"key":"S0218126624502712BIB020","first-page":"129","volume-title":"International Symposium on Field Programmable Gate Arrays (FPGA)","author":"Wang S.","year":"2020"},{"key":"S0218126624502712BIB021","first-page":"21","volume-title":"International Symposium on Field Programmable Gate Arrays (FPGA)","author":"Gao C.","year":"2020"},{"key":"S0218126624502712BIB023","doi-asserted-by":"publisher","DOI":"10.1145\/3392717.3392749"},{"key":"S0218126624502712BIB024","doi-asserted-by":"publisher","DOI":"10.1201\/9781003162810-13"},{"key":"S0218126624502712BIB028","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3117837"},{"key":"S0218126624502712BIB029","first-page":"3301","volume-title":"The Thirty-Sixth International Conference on Machine Learning (ICML)","author":"Kaya Y.","year":"2019"},{"key":"S0218126624502712BIB030","doi-asserted-by":"publisher","DOI":"10.1145\/3097983.3098177"},{"key":"S0218126624502712BIB033","first-page":"694","volume-title":"ECML PKDD","author":"Lambert A.","year":"2020"},{"key":"S0218126624502712BIB034","first-page":"6000","volume-title":"The Thirty-First Annual Conference on Neural Information Processing Systems (NIPS)","volume":"30","author":"Vaswani A.","year":"2017"}],"container-title":["Journal of Circuits, Systems and Computers"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.worldscientific.com\/doi\/pdf\/10.1142\/S0218126624502712","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,10,23]],"date-time":"2024-10-23T08:57:32Z","timestamp":1729673852000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.worldscientific.com\/doi\/10.1142\/S0218126624502712"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,6,10]]},"references-count":25,"journal-issue":{"issue":"15","published-print":{"date-parts":[[2024,10]]}},"alternative-id":["10.1142\/S0218126624502712"],"URL":"https:\/\/doi.org\/10.1142\/s0218126624502712","relation":{},"ISSN":["0218-1266","1793-6454"],"issn-type":[{"type":"print","value":"0218-1266"},{"type":"electronic","value":"1793-6454"}],"subject":[],"published":{"date-parts":[[2024,6,10]]},"article-number":"2450271"}}