{"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/variational-dropout-sparsifies-deep-neural","title":"Variational Dropout Sparsifies Deep Neural Networks","arxiv_id":"1701.05369","date":"2017-01-19","proceeding":"ICML 2017 8","authors":["Dmitry Molchanov","Arsenii Ashukha","Dmitry Vetrov"],"abstract":"We explore a recently proposed Variational Dropout technique that provided an\nelegant Bayesian interpretation to Gaussian Dropout. We extend Variational\nDropout to the case when dropout rates are unbounded, propose a way to reduce\nthe variance of the gradient estimator and report first experimental results\nwith individual dropout rates per weight. Interestingly, it leads to extremely\nsparse solutions both in fully-connected and convolutional layers. This effect\nis similar to automatic relevance determination effect in empirical Bayes but\nhas a number of advantages. We reduce the number of parameters up to 280 times\non LeNet architectures and up to 68 times on VGG-like networks with a\nnegligible decrease of accuracy.","url_abs":"http://arxiv.org/abs/1701.05369v3","url_pdf":"http://arxiv.org/pdf/1701.05369v3.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":"variational-dropout-sparsifies-deep-neural","repo_url":"https://github.com/Cerphilly/Sparse_VD_tf2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"variational-dropout-sparsifies-deep-neural","repo_url":"https://github.com/Faptimus420/Sparse_VD_keras-core","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"jax","reach":{"status":"ok"}},{"paper_slug":"variational-dropout-sparsifies-deep-neural","repo_url":"https://github.com/HolyBayes/VarDropPytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"variational-dropout-sparsifies-deep-neural","repo_url":"https://github.com/HolyBayes/pytorch_ard","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"variational-dropout-sparsifies-deep-neural","repo_url":"https://github.com/KarenUllrich/Tutorial_BayesianCompressionForDL","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"variational-dropout-sparsifies-deep-neural","repo_url":"https://github.com/Leensman/VarDropPytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"variational-dropout-sparsifies-deep-neural","repo_url":"https://github.com/MSergeyV/Adaptive-dropout-with-Rademacher-complexity","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"variational-dropout-sparsifies-deep-neural","repo_url":"https://github.com/ModelZoos/ModelZooDataset","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"variational-dropout-sparsifies-deep-neural","repo_url":"https://github.com/ars-ashuha/sparse-vd-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"variational-dropout-sparsifies-deep-neural","repo_url":"https://github.com/ars-ashuha/variational-dropout-sparsifies-dnn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"GPL-3.0"}},{"paper_slug":"variational-dropout-sparsifies-deep-neural","repo_url":"https://github.com/cbbjames/Variational-Dropout---ResNet-","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"GPL-3.0"}},{"paper_slug":"variational-dropout-sparsifies-deep-neural","repo_url":"https://github.com/maxblumental/variational-drouput","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"variational-dropout-sparsifies-deep-neural","repo_url":"https://github.com/senya-ashukha/sparse-vd-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"variational-dropout-sparsifies-deep-neural","repo_url":"https://github.com/senya-ashukha/variational-dropout-sparsifies-dnn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"GPL-3.0"}},{"paper_slug":"variational-dropout-sparsifies-deep-neural","repo_url":"https://github.com/xaosina/Struct-Sparse-Pruning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"sparse-learning","task_name":"Sparse Learning"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"lenet","method_name":"LeNet"},{"method_slug":"variational-dropout","method_name":"Variational Dropout"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1701.05369","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1701.05369"}},"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":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ars-ashuha/sparse-vd-pytorch","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ars-ashuha/variational-dropout-sparsifies-dnn","reach":{"status":"ok","spdx":"GPL-3.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Cerphilly/Sparse_VD_tf2","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/HolyBayes/VarDropPytorch","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/xaosina/Struct-Sparse-Pruning","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/KarenUllrich/Tutorial_BayesianCompressionForDL","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/HolyBayes/pytorch_ard","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Faptimus420/Sparse_VD_keras-core","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/senya-ashukha/variational-dropout-sparsifies-dnn","reach":{"status":"ok","spdx":"GPL-3.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ModelZoos/ModelZooDataset","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/maxblumental/variational-drouput","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/cbbjames/Variational-Dropout---ResNet-","reach":{"status":"ok","spdx":"GPL-3.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Leensman/VarDropPytorch","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/senya-ashukha/sparse-vd-pytorch","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/MSergeyV/Adaptive-dropout-with-Rademacher-complexity","reach":{"status":"ok"}}],"summary":{"ran_honours":1,"unverified":2},"by_repo_kind":{"listed":{"samples":3,"ran":1,"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":"ddb40d6a6a8ec24f","entry":"get_kl_weight","repo":"HolyBayes/pytorch_ard","repo_kind":"listed","path":"examples/boston/boston_ard.py","file_url":"https://github.com/HolyBayes/pytorch_ard/blob/HEAD/examples/boston/boston_ard.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ddb40d6a6a8ec24f"}},{"code_sha256_prefix":"b4e9bbfcc65bddb1","entry":"get_ard_reg","repo":"HolyBayes/pytorch_ard","repo_kind":"listed","path":"torch_ard/torch_ard.py","file_url":"https://github.com/HolyBayes/pytorch_ard/blob/HEAD/torch_ard/torch_ard.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b4e9bbfcc65bddb1"}},{"code_sha256_prefix":"af3187f43f548f04","entry":"get_dropped_params_ratio","repo":"HolyBayes/pytorch_ard","repo_kind":"listed","path":"torch_ard/torch_ard.py","file_url":"https://github.com/HolyBayes/pytorch_ard/blob/HEAD/torch_ard/torch_ard.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"af3187f43f548f04"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}