{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/to-prune-or-not-to-prune-exploring-the","title":"To prune, or not to prune: exploring the efficacy of pruning for model compression","arxiv_id":"1710.01878","date":"2017-10-05","proceeding":"ICLR 2018 1","authors":["Michael Zhu","Suyog Gupta"],"abstract":"Model pruning seeks to induce sparsity in a deep neural network's various\nconnection matrices, thereby reducing the number of nonzero-valued parameters\nin the model. Recent reports (Han et al., 2015; Narang et al., 2017) prune deep\nnetworks at the cost of only a marginal loss in accuracy and achieve a sizable\nreduction in model size. This hints at the possibility that the baseline models\nin these experiments are perhaps severely over-parameterized at the outset and\na viable alternative for model compression might be to simply reduce the number\nof hidden units while maintaining the model's dense connection structure,\nexposing a similar trade-off in model size and accuracy. We investigate these\ntwo distinct paths for model compression within the context of energy-efficient\ninference in resource-constrained environments and propose a new gradual\npruning technique that is simple and straightforward to apply across a variety\nof models/datasets with minimal tuning and can be seamlessly incorporated\nwithin the training process. We compare the accuracy of large, but pruned\nmodels (large-sparse) and their smaller, but dense (small-dense) counterparts\nwith identical memory footprint. Across a broad range of neural network\narchitectures (deep CNNs, stacked LSTM, and seq2seq LSTM models), we find\nlarge-sparse models to consistently outperform small-dense models and achieve\nup to 10x reduction in number of non-zero parameters with minimal loss in\naccuracy.","url_abs":"http://arxiv.org/abs/1710.01878v2","url_pdf":"http://arxiv.org/pdf/1710.01878v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"to-prune-or-not-to-prune-exploring-the","repo_url":"https://github.com/RAIVNLab/STR","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"to-prune-or-not-to-prune-exploring-the","repo_url":"https://github.com/dorlivne/simple_net_pruning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"to-prune-or-not-to-prune-exploring-the","repo_url":"https://github.com/intellabs/model-compression-research-package","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"gone","observed_at":"2026-09-17","how":"tree_404+repo_404"}},{"paper_slug":"to-prune-or-not-to-prune-exploring-the","repo_url":"https://github.com/klieberman/ood_nic","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"model-compression","task_name":"Model Compression"}],"methods":[{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"pruning","method_name":"Pruning"},{"method_slug":"seq2seq","method_name":"Seq2Seq"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1710.01878","atlas_url":"https://app.syntology.ai/?focus=1710.01878","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1710.01878"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/RAIVNLab/STR","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/klieberman/ood_nic","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/intellabs/model-compression-research-package","reach":{"status":"gone","observed_at":"2026-09-17","how":"tree_404+repo_404"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/dorlivne/simple_net_pruning","reach":{"status":"ok"}}],"summary":{"ran_draft_wrong":2},"by_repo_kind":{"listed":{"samples":2,"ran":2,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"892023030e436d8c","entry":"GMPChooseEdges","repo":"RAIVNLab/STR","repo_kind":"listed","path":"utils/conv_type.py","file_url":"https://github.com/RAIVNLab/STR/blob/HEAD/utils/conv_type.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"892023030e436d8c"}},{"code_sha256_prefix":"d36848c6cabd141e","entry":"sparseFunction","repo":"RAIVNLab/STR","repo_kind":"listed","path":"utils/conv_type.py","file_url":"https://github.com/RAIVNLab/STR/blob/HEAD/utils/conv_type.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d36848c6cabd141e"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}