{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T16:23:38Z","timestamp":1773246218691,"version":"3.50.1"},"reference-count":44,"publisher":"Association for Computing Machinery (ACM)","issue":"6","license":[{"start":{"date-parts":[[2024,9,11]],"date-time":"2024-09-11T00:00:00Z","timestamp":1726012800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"DST-INRIA-CNRS","award":["IFC\/4131\/DST-Inria\/2018-2019\/1"],"award-info":[{"award-number":["IFC\/4131\/DST-Inria\/2018-2019\/1"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Embed. Comput. Syst."],"published-print":{"date-parts":[[2024,11,30]]},"abstract":"<jats:p>Neural network model compression is very important to achieve model deployment based on the memory and storage available in different computing systems. Generally, the continuous drive for higher accuracy in these models increases their size and complexity, making it challenging to deploy them on resource-constrained computing environments. This article proposes various algorithms for model compression by exploiting weight characteristics and conducts an in-depth study of their performance. The algorithms involve manipulating exponents and mantissa in the floating-point representations of weights. In addition, we also present a retraining method that uses the proposed algorithms to further reduce the size of pre-trained models. The results presented in this article are mainly on BFloat16 floating-point format. The proposed weight manipulation algorithms save at least 20% of memory on state-of-the-art image classification models with very minor accuracy loss. This loss is bridged using the retraining method that saves at least 30% of memory, with potential memory savings of up to 43%. We compare the performance of the proposed methods against the state-of-the-art model compression techniques in terms of accuracy, memory savings, inference time, and energy.<\/jats:p>","DOI":"10.1145\/3687466","type":"journal-article","created":{"date-parts":[[2024,8,10]],"date-time":"2024-08-10T11:20:49Z","timestamp":1723288849000},"page":"1-23","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Combining Weight Approximation, Sharing and Retraining for Neural Network Model Compression"],"prefix":"10.1145","volume":"23","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2280-2990","authenticated-orcid":false,"given":"Prachi","family":"Kashikar","sequence":"first","affiliation":[{"name":"Indian Institute of Technology Goa, Ponda, India"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4334-6418","authenticated-orcid":false,"given":"Olivier","family":"Sentieys","sequence":"additional","affiliation":[{"name":"INRIA, University of Rennes, Rennes, France"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4532-2017","authenticated-orcid":false,"given":"Sharad","family":"Sinha","sequence":"additional","affiliation":[{"name":"Computer Science, IIT Goa, Ponda, India"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2024,9,11]]},"reference":[{"key":"e_1_3_1_2_2","unstructured":"Xilinx. 2018. Zynq-7000 SoC Data Sheet: Overview. Retrieved February 28 2023 from https:\/\/www.mouser.com\/datasheet\/2\/903\/ds190-Zynq-7000-Overview-1595492.pdf"},{"key":"e_1_3_1_3_2","first-page":"265","volume-title":"Proceedings of the 12th USENIX Symposium on Operating Systems Design and Implementation (OSDI \u201916)","author":"Abadi Mart\u00edn","year":"2016","unstructured":"Mart\u00edn Abadi, Paul Barham, Jianmin Chen, Zhifeng Chen, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Geoffrey Irving, Michael Isard, Manjunath Kudlur, Josh Levenberg, Rajat Monga, Sherry Moore, Derek G. Murray, Benoit Steiner, Paul Tucker, Vijay Vasudevan, Pete Warden, Martin Wicke, Yuan Yu, and Xiaoqiang Zheng. 2016. TensorFlow: A system for large-scale machine learning. In Proceedings of the 12th USENIX Symposium on Operating Systems Design and Implementation (OSDI \u201916). 265\u2013283. https:\/\/www.usenix.org\/conference\/osdi16\/technical-sessions\/presentation\/abadi"},{"key":"e_1_3_1_4_2","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2018.2874767"},{"key":"e_1_3_1_5_2","first-page":"8532","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"Carreira-Perpin\u00e1n Miguel A.","year":"2018","unstructured":"Miguel A. Carreira-Perpin\u00e1n and Yerlan Idelbayev. 2018. \u201cLearning-compression\u201d algorithms for neural net pruning. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 8532\u20138541."},{"key":"e_1_3_1_6_2","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA51647.2021.00027"},{"key":"e_1_3_1_7_2","unstructured":"Hsu-Hsun Chin Ren-Song Tsay and Hsin-I. Wu. 2021. A high-performance adaptive quantization approach for edge CNN applications. arXiv:2107.08382[cs.CV] (2021)."},{"key":"e_1_3_1_8_2","doi-asserted-by":"publisher","DOI":"10.1145\/178243.178259"},{"key":"e_1_3_1_9_2","doi-asserted-by":"publisher","DOI":"10.23919\/DATE48585.2020.9116350"},{"key":"e_1_3_1_10_2","doi-asserted-by":"publisher","DOI":"10.1109\/DDECS50862.2020.9095658"},{"key":"e_1_3_1_11_2","article-title":"Lottery tickets in linear models: An analysis of iterative magnitude pruning","author":"Elesedy Bryn","year":"2020","unstructured":"Bryn Elesedy, Varun Kanade, and Yee Whye Teh. 2020. Lottery tickets in linear models: An analysis of iterative magnitude pruning. arXiv preprint arXiv:2007.08243 (2020).","journal-title":"arXiv preprint arXiv:2007.08243"},{"key":"e_1_3_1_12_2","doi-asserted-by":"publisher","DOI":"10.1109\/TIE.2021.3070517"},{"key":"e_1_3_1_13_2","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2020.2996231"},{"key":"e_1_3_1_14_2","doi-asserted-by":"publisher","DOI":"10.1145\/3487045"},{"key":"e_1_3_1_15_2","volume-title":"Advances in Neural Information Processing Systems","author":"Han Song","year":"2015","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, C. Cortes, N. Lawrence, D. Lee, M. Sugiyama, and R. Garnett (Eds.). Vol. 28. Curran Associates, 1\u20139.https:\/\/proceedings.neurips.cc\/paper_files\/paper\/2015\/file\/ae0eb3eed39d2bcef4622b2499a05fe6-Paper.pdf"},{"key":"e_1_3_1_16_2","unstructured":"Kaiming He Xiangyu Zhang Shaoqing Ren and Jian Sun. 2015. Deep residual learning for image recognition. arXiv:1512.03385[cs.CV] (2015)."},{"key":"e_1_3_1_17_2","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2020.3040367"},{"key":"e_1_3_1_18_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00140"},{"key":"e_1_3_1_19_2","doi-asserted-by":"crossref","unstructured":"Gao Huang Zhuang Liu Laurens van der Maaten and Kilian Q. Weinberger. 2018. Densely connected convolutional networks. arXiv:1608.06993[cs.CV] (2018).","DOI":"10.1109\/CVPR.2017.243"},{"key":"e_1_3_1_20_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2010.107"},{"key":"e_1_3_1_21_2","unstructured":"Dhiraj Kalamkar Dheevatsa Mudigere Naveen Mellempudi Dipankar Das Kunal Banerjee Sasikanth Avancha Dharma Teja Vooturi Nataraj Jammalamadaka Jianyu Huang Hector Yuen Jiyan Yang Jongsoo Park Alexander Heinecke Evangelos Georganas Sudarshan Srinivasan Abhisek Kundu Misha Smelyanskiy Bharat Kaul and Pradeep Dubey. 2019. A study of BFLOAT16 for deep learning training. arXiv:1905.12322[cs.LG] (2019)."},{"key":"e_1_3_1_22_2","doi-asserted-by":"publisher","DOI":"10.1109\/TVLSI.2023.3307607"},{"key":"e_1_3_1_23_2","doi-asserted-by":"publisher","DOI":"10.1109\/AICAS51828.2021.9458581"},{"key":"e_1_3_1_24_2","volume-title":"Learning Multiple Layers of Features from Tiny Images","author":"Krizhevsky Alex","year":"2009","unstructured":"Alex Krizhevsky. 2009. Learning Multiple Layers of Features from Tiny Images. Technical Report. University of Toronto."},{"key":"e_1_3_1_25_2","article-title":"Deep convolutional neural network inference with floating-point weights and fixed-point activations","author":"Lai Liangzhen","year":"2017","unstructured":"Liangzhen Lai, Naveen Suda, and Vikas Chandra. 2017. Deep convolutional neural network inference with floating-point weights and fixed-point activations. arXiv preprint arXiv:1703.03073 (2017).","journal-title":"arXiv preprint arXiv:1703.03073"},{"key":"e_1_3_1_26_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3066410"},{"key":"e_1_3_1_27_2","article-title":"Data-free knowledge distillation for deep neural networks","author":"Lopes Raphael Gontijo","year":"2017","unstructured":"Raphael Gontijo Lopes, Stefano Fenu, and Thad Starner. 2017. Data-free knowledge distillation for deep neural networks. arXiv preprint arXiv:1710.07535 (2017).","journal-title":"arXiv preprint arXiv:1710.07535"},{"key":"e_1_3_1_28_2","doi-asserted-by":"crossref","unstructured":"Sashank Macha Om Oza Alex Escott Francesco Caliva Robbie Armitano Santosh Kumar Cheekatmalla Sree Hari Krishnan Parthasarathi and Yuzong Liu. 2023. Fixed-point quantization aware training for on-device keyword-spotting. arXiv:2303.02284[eess.AS] (2023).","DOI":"10.1109\/ICASSP49357.2023.10095977"},{"key":"e_1_3_1_29_2","first-page":"17629","article-title":"Pruning filter in filter","volume":"33","author":"Meng Fanxu","year":"2020","unstructured":"Fanxu Meng, Hao Cheng, Ke Li, Huixiang Luo, Xiaowei Guo, Guangming Lu, and Xing Sun. 2020. Pruning filter in filter. Advances in Neural Information Processing Systems 33 (2020), 17629\u201317640.","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_3_1_30_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICMLA.2019.00127"},{"key":"e_1_3_1_31_2","unstructured":"Adam Paszke Sam Gross Francisco Massa Adam Lerer James Bradbury Gregory Chanan Trevor Killeen Zeming Lin Natalia Gimelshein Luca Antiga Alban Desmaison Andreas K\u00f6pf Edward Yang Zach DeVito Martin Raison Alykhan Tejani Sasank Chilamkurthy Benoit Steiner Lu Fang Junjie Bai and Soumith Chintala. 2019. PyTorch: An imperative style high-performance deep learning library. arXiv:1912.01703[cs.LG] (2019)."},{"key":"e_1_3_1_32_2","doi-asserted-by":"publisher","DOI":"10.1109\/ATEE52255.2021.9425265"},{"key":"e_1_3_1_33_2","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.4431043"},{"key":"e_1_3_1_34_2","doi-asserted-by":"publisher","DOI":"10.1145\/3410463.3414648"},{"key":"e_1_3_1_35_2","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2023.3258360"},{"key":"e_1_3_1_36_2","unstructured":"Victor Schmidt Kamal Goyal Aditya Joshi Boris Feld Liam Conell Nikolas Laskaris Doug Blank Jonathan Wilson Sorelle Friedler and Sasha Luccioni. 2021. CodeCarbon: Estimate and Track Carbon Emissions from Machine Learning Computing. Retrieved August 13 2024 from https:\/\/codecarbon.io"},{"key":"e_1_3_1_37_2","unstructured":"Karen Simonyan and Andrew Zisserman. 2015. Very deep convolutional networks for large-scale image recognition. arXiv:1409.1556[cs.CV] (2015)."},{"key":"e_1_3_1_38_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"e_1_3_1_39_2","article-title":"Soft weight-sharing for neural network compression","author":"Ullrich Karen","year":"2017","unstructured":"Karen Ullrich, Edward Meeds, and Max Welling. 2017. Soft weight-sharing for neural network compression. arXiv preprint arXiv:1702.04008 (2017).","journal-title":"arXiv preprint arXiv:1702.04008"},{"key":"e_1_3_1_40_2","doi-asserted-by":"publisher","DOI":"10.1109\/TCSII.2022.3224022"},{"key":"e_1_3_1_41_2","doi-asserted-by":"publisher","DOI":"10.1145\/3626100"},{"key":"e_1_3_1_42_2","doi-asserted-by":"publisher","DOI":"10.1109\/SSCI50451.2021.9659887"},{"key":"e_1_3_1_43_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746762"},{"key":"e_1_3_1_44_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.643"},{"key":"e_1_3_1_45_2","article-title":"To prune, or not to prune: Exploring the efficacy of pruning for model compression","author":"Zhu Michael","year":"2017","unstructured":"Michael Zhu and Suyog Gupta. 2017. To prune, or not to prune: Exploring the efficacy of pruning for model compression. arXiv preprint arXiv:1710.01878 (2017).","journal-title":"arXiv preprint arXiv:1710.01878"}],"container-title":["ACM Transactions on Embedded Computing Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3687466","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3687466","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T00:05:59Z","timestamp":1750291559000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3687466"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,9,11]]},"references-count":44,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2024,11,30]]}},"alternative-id":["10.1145\/3687466"],"URL":"https:\/\/doi.org\/10.1145\/3687466","relation":{},"ISSN":["1539-9087","1558-3465"],"issn-type":[{"value":"1539-9087","type":"print"},{"value":"1558-3465","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,9,11]]},"assertion":[{"value":"2024-02-23","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-08-01","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-09-11","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}