{"_id":"10180","language":[{"iso":"eng"}],"citation":{"chicago":"Hoefler, Torsten, Dan-Adrian Alistarh, Tal Ben-Nun, Nikoli Dryden, and Elena-Alexandra Peste. “Sparsity in Deep Learning: Pruning and Growth for Efficient Inference and Training in Neural Networks.” <i>Journal of Machine Learning Research</i>. Journal of Machine Learning Research, 2021.","apa":"Hoefler, T., Alistarh, D.-A., Ben-Nun, T., Dryden, N., &#38; Peste, E.-A. (2021). Sparsity in deep learning: Pruning and growth for efficient inference and training in neural networks. <i>Journal of Machine Learning Research</i>. Journal of Machine Learning Research.","ista":"Hoefler T, Alistarh D-A, Ben-Nun T, Dryden N, Peste E-A. 2021. Sparsity in deep learning: Pruning and growth for efficient inference and training in neural networks. Journal of Machine Learning Research. 22(241), 1–124.","ama":"Hoefler T, Alistarh D-A, Ben-Nun T, Dryden N, Peste E-A. Sparsity in deep learning: Pruning and growth for efficient inference and training in neural networks. <i>Journal of Machine Learning Research</i>. 2021;22(241):1-124.","short":"T. Hoefler, D.-A. Alistarh, T. Ben-Nun, N. Dryden, E.-A. Peste, Journal of Machine Learning Research 22 (2021) 1–124.","mla":"Hoefler, Torsten, et al. “Sparsity in Deep Learning: Pruning and Growth for Efficient Inference and Training in Neural Networks.” <i>Journal of Machine Learning Research</i>, vol. 22, no. 241, Journal of Machine Learning Research, 2021, pp. 1–124.","ieee":"T. Hoefler, D.-A. Alistarh, T. Ben-Nun, N. Dryden, and E.-A. Peste, “Sparsity in deep learning: Pruning and growth for efficient inference and training in neural networks,” <i>Journal of Machine Learning Research</i>, vol. 22, no. 241. Journal of Machine Learning Research, pp. 1–124, 2021."},"article_type":"original","external_id":{"arxiv":["2102.00554"]},"publication_identifier":{"eissn":["1533-7928"],"issn":["1532-4435"]},"author":[{"last_name":"Hoefler","first_name":"Torsten","full_name":"Hoefler, Torsten"},{"orcid":"0000-0003-3650-940X","id":"4A899BFC-F248-11E8-B48F-1D18A9856A87","full_name":"Alistarh, Dan-Adrian","first_name":"Dan-Adrian","last_name":"Alistarh"},{"full_name":"Ben-Nun, Tal","last_name":"Ben-Nun","first_name":"Tal"},{"full_name":"Dryden, Nikoli","last_name":"Dryden","first_name":"Nikoli"},{"first_name":"Elena-Alexandra","last_name":"Peste","full_name":"Peste, Elena-Alexandra","id":"32D78294-F248-11E8-B48F-1D18A9856A87"}],"has_accepted_license":"1","publication":"Journal of Machine Learning Research","page":"1-124","ddc":["000"],"acknowledgement":"We thank Doug Burger, Steve Scott, Marco Heddes, and the respective teams at Microsoft for inspiring discussions on the topic. We thank Angelika Steger for uplifting debates about the connections to biological brains, Sidak Pal Singh for his support regarding experimental results, and Utku Evci as well as Xin Wang for comments on previous versions of this\r\nwork. Special thanks go to Bernhard Schölkopf, our JMLR editor Samy Bengio, and the three anonymous reviewers who provided excellent comprehensive, pointed, and deep review comments that improved the quality of our manuscript significantly.","file_date_updated":"2021-10-27T15:34:18Z","year":"2021","intvolume":"        22","volume":22,"department":[{"_id":"DaAl"}],"abstract":[{"text":"The growing energy and performance costs of deep learning have driven the community to reduce the size of neural networks by selectively pruning components. Similarly to their biological counterparts, sparse networks generalize just as well, sometimes even better than, the original dense networks. Sparsity promises to reduce the memory footprint of regular networks to fit mobile devices, as well as shorten training time for ever growing networks. In this paper, we survey prior work on sparsity in deep learning and provide an extensive tutorial of sparsification for both inference and training. We describe approaches to remove and add elements of neural networks, different training strategies to achieve model sparsity, and mechanisms to exploit sparsity in practice. Our work distills ideas from more than 300 research papers and provides guidance to practitioners who wish to utilize sparsity today, as well as to researchers whose goal is to push the frontier forward. We include the necessary background on mathematical methods in sparsification, describe phenomena such as early structure adaptation, the intricate relations between sparsity and the training process, and show techniques for achieving acceleration on real hardware. We also define a metric of pruned parameter efficiency that could serve as a baseline for comparison of different sparse networks. We close by speculating on how sparsity can improve future workloads and outline major open problems in the field.","lang":"eng"}],"main_file_link":[{"open_access":"1","url":"https://www.jmlr.org/papers/v22/21-0366.html"}],"date_published":"2021-09-01T00:00:00Z","publication_status":"published","issue":"241","article_processing_charge":"No","month":"09","date_updated":"2022-05-13T09:36:08Z","tmp":{"image":"/images/cc_by.png","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","short":"CC BY (4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode"},"type":"journal_article","file":[{"file_name":"2021_JMachLearnRes_Hoefler.pdf","creator":"cziletti","access_level":"open_access","content_type":"application/pdf","relation":"main_file","checksum":"3389d9d01fc58f8fb4c1a53e14a8abbf","date_created":"2021-10-27T15:34:18Z","file_size":3527521,"file_id":"10192","success":1,"date_updated":"2021-10-27T15:34:18Z"}],"day":"01","title":"Sparsity in deep learning: Pruning and growth for efficient inference and training in neural networks","quality_controlled":"1","status":"public","publisher":"Journal of Machine Learning Research","scopus_import":"1","oa_version":"Published Version","oa":1,"date_created":"2021-10-24T22:01:34Z","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87"}