{"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/coupled-ensembles-of-neural-networks","title":"Coupled Ensembles of Neural Networks","arxiv_id":"1709.06053","date":"2017-09-18","proceeding":"ICLR 2018 1","authors":["Anuvabh Dutt","Denis Pellerin","Georges Quénot"],"abstract":"We investigate in this paper the architecture of deep convolutional networks.\nBuilding on existing state of the art models, we propose a reconfiguration of\nthe model parameters into several parallel branches at the global network\nlevel, with each branch being a standalone CNN. We show that this arrangement\nis an efficient way to significantly reduce the number of parameters without\nlosing performance or to significantly improve the performance with the same\nlevel of performance. The use of branches brings an additional form of\nregularization. In addition to the split into parallel branches, we propose a\ntighter coupling of these branches by placing the \"fuse (averaging) layer\"\nbefore the Log-Likelihood and SoftMax layers during training. This gives\nanother significant performance improvement, the tighter coupling favouring the\nlearning of better representations, even at the level of the individual\nbranches. We refer to this branched architecture as \"coupled ensembles\". The\napproach is very generic and can be applied with almost any DCNN architecture.\nWith coupled ensembles of DenseNet-BC and parameter budget of 25M, we obtain\nerror rates of 2.92%, 15.68% and 1.50% respectively on CIFAR-10, CIFAR-100 and\nSVHN tasks. For the same budget, DenseNet-BC has error rate of 3.46%, 17.18%,\nand 1.8% respectively. With ensembles of coupled ensembles, of DenseNet-BC\nnetworks, with 50M total parameters, we obtain error rates of 2.72%, 15.13% and\n1.42% respectively on these tasks.","url_abs":"http://arxiv.org/abs/1709.06053v1","url_pdf":"http://arxiv.org/pdf/1709.06053v1.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":"coupled-ensembles-of-neural-networks","repo_url":"https://github.com/vabh/coupled_ensembles","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"coupled-ensembles-of-neural-networks","repo_url":"https://github.com/grey-area/modular-loss-experiments","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[],"methods":[{"method_slug":"dcnn","method_name":"DCNN"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1709.06053","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1709.06053"}},"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/grey-area/modular-loss-experiments","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/vabh/coupled_ensembles","reach":null}],"summary":{"ran_honours":1},"by_repo_kind":{"official":{"samples":1,"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":1,"samples":[{"code_sha256_prefix":"89be6edf313246b5","entry":"train","repo":"vabh/coupled_ensembles","repo_kind":"official","path":"train_model.py","file_url":"https://github.com/vabh/coupled_ensembles/blob/HEAD/train_model.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"89be6edf313246b5"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}