{"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/binaryconnect-training-deep-neural-networks","title":"BinaryConnect: Training Deep Neural Networks with binary weights during propagations","arxiv_id":"1511.00363","date":"2015-11-02","proceeding":"NeurIPS 2015 12","authors":["Matthieu Courbariaux","Yoshua Bengio","Jean-Pierre David"],"abstract":"Deep Neural Networks (DNN) have achieved state-of-the-art results in a wide\nrange of tasks, with the best results obtained with large training sets and\nlarge models. In the past, GPUs enabled these breakthroughs because of their\ngreater computational speed. In the future, faster computation at both training\nand test time is likely to be crucial for further progress and for consumer\napplications on low-power devices. As a result, there is much interest in\nresearch and development of dedicated hardware for Deep Learning (DL). Binary\nweights, i.e., weights which are constrained to only two possible values (e.g.\n-1 or 1), would bring great benefits to specialized DL hardware by replacing\nmany multiply-accumulate operations by simple accumulations, as multipliers are\nthe most space and power-hungry components of the digital implementation of\nneural networks. We introduce BinaryConnect, a method which consists in\ntraining a DNN with binary weights during the forward and backward\npropagations, while retaining precision of the stored weights in which\ngradients are accumulated. Like other dropout schemes, we show that\nBinaryConnect acts as regularizer and we obtain near state-of-the-art results\nwith BinaryConnect on the permutation-invariant MNIST, CIFAR-10 and SVHN.","url_abs":"http://arxiv.org/abs/1511.00363v3","url_pdf":"http://arxiv.org/pdf/1511.00363v3.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":"binaryconnect-training-deep-neural-networks","repo_url":"https://github.com/MatthieuCourbariaux/BinaryConnect","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"GPL-2.0"}},{"paper_slug":"binaryconnect-training-deep-neural-networks","repo_url":"https://github.com/Akaza994/Binary-Quantization","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"binaryconnect-training-deep-neural-networks","repo_url":"https://github.com/lzj994/Binary-Quantization","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"binaryconnect-training-deep-neural-networks","repo_url":"https://github.com/ryuz/BinaryBrain","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"binaryconnect-training-deep-neural-networks","repo_url":"https://github.com/tensorpack/tensorpack/tree/master/examples/DoReFa-Net","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[],"methods":[{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"hard-sigmoid","method_name":"Hard Sigmoid"}],"datasets_introduced":[],"methods_introduced":[{"slug":"hard-sigmoid","name":"Hard Sigmoid","full_name":"Hard Sigmoid"}],"results":[{"leaderboard":"/sota/image-classification-on-cifar-10","task":"Image Classification","dataset":"CIFAR-10","model":"BinaryConnect","rank_in_archive_order":189,"of":265,"metrics":{"Percentage correct":"91.7"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-mnist","task":"Image Classification","dataset":"MNIST","model":"BinaryConnect","rank_in_archive_order":47,"of":81,"metrics":{"Percentage error":"1.0"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-svhn","task":"Image Classification","dataset":"SVHN","model":"BinaryConnect","rank_in_archive_order":30,"of":62,"metrics":{"Percentage error":"2.2"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1511.00363","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1511.00363"}},"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/Akaza994/Binary-Quantization","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/MatthieuCourbariaux/BinaryConnect","reach":{"status":"ok","spdx":"GPL-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/lzj994/Binary-Quantization","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/tensorpack/tensorpack/tree/master/examples/DoReFa-Net","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ryuz/BinaryBrain","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":2,"ran_fixture":1},"by_repo_kind":{},"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":3,"samples":[{"code_sha256_prefix":"03a0be4fb7381bd2","entry":"conv3x3","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"03a0be4fb7381bd2"}},{"code_sha256_prefix":"c9385e42b40d1dea","entry":"get_model2","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"c9385e42b40d1dea"}},{"code_sha256_prefix":"cff17e6f91a48ebd","entry":"quantize_bw","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"cff17e6f91a48ebd"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}