{"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/dynamic-structure-pruning-for-compressing","title":"Dynamic Structure Pruning for Compressing CNNs","arxiv_id":"2303.09736","date":"2023-03-17","proceeding":null,"authors":["Jun-Hyung Park","Yeachan Kim","Junho Kim","Joon-Young Choi","SangKeun Lee"],"abstract":"Structure pruning is an effective method to compress and accelerate neural networks. While filter and channel pruning are preferable to other structure pruning methods in terms of realistic acceleration and hardware compatibility, pruning methods with a finer granularity, such as intra-channel pruning, are expected to be capable of yielding more compact and computationally efficient networks. Typical intra-channel pruning methods utilize a static and hand-crafted pruning granularity due to a large search space, which leaves room for improvement in their pruning performance. In this work, we introduce a novel structure pruning method, termed as dynamic structure pruning, to identify optimal pruning granularities for intra-channel pruning. In contrast to existing intra-channel pruning methods, the proposed method automatically optimizes dynamic pruning granularities in each layer while training deep neural networks. To achieve this, we propose a differentiable group learning method designed to efficiently learn a pruning granularity based on gradient-based learning of filter groups. The experimental results show that dynamic structure pruning achieves state-of-the-art pruning performance and better realistic acceleration on a GPU compared with channel pruning. In particular, it reduces the FLOPs of ResNet50 by 71.85% without accuracy degradation on the ImageNet dataset. Our code is available at https://github.com/irishev/DSP.","url_abs":"https://arxiv.org/abs/2303.09736v1","url_pdf":"https://arxiv.org/pdf/2303.09736v1.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":"dynamic-structure-pruning-for-compressing","repo_url":"https://github.com/irishev/dsp","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":null,"task_name":"GPU"}],"methods":[{"method_slug":"pruning","method_name":"Pruning"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2303.09736","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2303.09736"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+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":"deterministic:regex_extraction","url":"https://github.com/pytorch/examples","reach":{"status":"ok","spdx":"BSD-3-Clause"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/irishev/DSP","reach":null}],"summary":{"ran":2,"unverified":6},"by_repo_kind":{"official":{"samples":2,"ran":2,"repositories":1},"found_in_text":{"samples":6,"ran":0,"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":"3cf9a1367bf5bda7","entry":"GroupWrapper","repo":"irishev/DSP","repo_kind":"official","path":"dsp_module.py","file_url":"https://github.com/irishev/DSP/blob/HEAD/dsp_module.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"3cf9a1367bf5bda7"}},{"code_sha256_prefix":"56118d0018b611ed","entry":"Hook","repo":"irishev/DSP","repo_kind":"official","path":"dsp_module.py","file_url":"https://github.com/irishev/DSP/blob/HEAD/dsp_module.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"56118d0018b611ed"}},{"code_sha256_prefix":"468aaad299edb475","entry":"apply_rotary_emb","repo":"pytorch/examples","repo_kind":"found_in_text","path":"distributed/tensor_parallelism/llama2_model.py","file_url":"https://github.com/pytorch/examples/blob/HEAD/distributed/tensor_parallelism/llama2_model.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"468aaad299edb475"}},{"code_sha256_prefix":"57154444b1a7a783","entry":"assert_all_nodes_have_tags","repo":"pytorch/examples","repo_kind":"found_in_text","path":"fx/custom_tracer.py","file_url":"https://github.com/pytorch/examples/blob/HEAD/fx/custom_tracer.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"57154444b1a7a783"}},{"code_sha256_prefix":"a2229a8c3bb45dd9","entry":"f","repo":"pytorch/examples","repo_kind":"found_in_text","path":"fx/invert.py","file_url":"https://github.com/pytorch/examples/blob/HEAD/fx/invert.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"a2229a8c3bb45dd9"}},{"code_sha256_prefix":"f3f961a13bd98c4b","entry":"invert","repo":"pytorch/examples","repo_kind":"found_in_text","path":"fx/invert.py","file_url":"https://github.com/pytorch/examples/blob/HEAD/fx/invert.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"f3f961a13bd98c4b"}},{"code_sha256_prefix":"ef236c5425b4e712","entry":"precompute_freqs_cis","repo":"pytorch/examples","repo_kind":"found_in_text","path":"distributed/tensor_parallelism/llama2_model.py","file_url":"https://github.com/pytorch/examples/blob/HEAD/distributed/tensor_parallelism/llama2_model.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"ef236c5425b4e712"}},{"code_sha256_prefix":"42d4e74777fd43b2","entry":"reshape_for_broadcast","repo":"pytorch/examples","repo_kind":"found_in_text","path":"distributed/tensor_parallelism/llama2_model.py","file_url":"https://github.com/pytorch/examples/blob/HEAD/distributed/tensor_parallelism/llama2_model.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"42d4e74777fd43b2"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}