{"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/fractional-max-pooling","title":"Fractional Max-Pooling","arxiv_id":"1412.6071","date":"2014-12-18","proceeding":null,"authors":["Benjamin Graham"],"abstract":"Convolutional networks almost always incorporate some form of spatial\npooling, and very often it is alpha times alpha max-pooling with alpha=2.\nMax-pooling act on the hidden layers of the network, reducing their size by an\ninteger multiplicative factor alpha. The amazing by-product of discarding 75%\nof your data is that you build into the network a degree of invariance with\nrespect to translations and elastic distortions. However, if you simply\nalternate convolutional layers with max-pooling layers, performance is limited\ndue to the rapid reduction in spatial size, and the disjoint nature of the\npooling regions. We have formulated a fractional version of max-pooling where\nalpha is allowed to take non-integer values. Our version of max-pooling is\nstochastic as there are lots of different ways of constructing suitable pooling\nregions. We find that our form of fractional max-pooling reduces overfitting on\na variety of datasets: for instance, we improve on the state-of-the art for\nCIFAR-100 without even using dropout.","url_abs":"http://arxiv.org/abs/1412.6071v4","url_pdf":"http://arxiv.org/pdf/1412.6071v4.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":"fractional-max-pooling","repo_url":"https://github.com/BendeguzToth/Fun-with-ConvNets","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"fractional-max-pooling","repo_url":"https://github.com/JonasWechsler/DeepLearningLab5","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"fractional-max-pooling","repo_url":"https://github.com/VladimirGol/KerasFractionalMaxPooling","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"fractional-max-pooling","repo_url":"https://github.com/facebookresearch/SparseConvNet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"fractional-max-pooling","repo_url":"https://github.com/laplacetw/vgg-like-cifar10","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-classification-on-cifar-10","task":"Image Classification","dataset":"CIFAR-10","model":"Fractional MP","rank_in_archive_order":108,"of":265,"metrics":{"Percentage correct":"96.5"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-cifar-100","task":"Image Classification","dataset":"CIFAR-100","model":"Fractional MP","rank_in_archive_order":161,"of":211,"metrics":{"Percentage correct":"73.6"},"uses_additional_data":true},{"leaderboard":"/sota/image-classification-on-mnist","task":"Image Classification","dataset":"MNIST","model":"Fractional MP","rank_in_archive_order":18,"of":81,"metrics":{"Percentage error":"0.3"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1412.6071","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1412.6071"}},"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/laplacetw/vgg-like-cifar10","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/BendeguzToth/Fun-with-ConvNets","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/VladimirGol/KerasFractionalMaxPooling","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/JonasWechsler/DeepLearningLab5","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/facebookresearch/SparseConvNet","reach":{"status":"ok","spdx":"NOASSERTION"}}],"summary":{"unverified":3},"by_repo_kind":{"listed":{"samples":3,"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":"191557b13a9f2a23","entry":"frac_max_pool","repo":"laplacetw/vgg-like-cifar10","repo_kind":"listed","path":"fmp_cifar10.py","file_url":"https://github.com/laplacetw/vgg-like-cifar10/blob/HEAD/fmp_cifar10.py","link_basis":"harvester_set","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":"191557b13a9f2a23"}},{"code_sha256_prefix":"e906d52e487c7b36","entry":"lr_decay","repo":"laplacetw/vgg-like-cifar10","repo_kind":"listed","path":"fmp_cifar10.py","file_url":"https://github.com/laplacetw/vgg-like-cifar10/blob/HEAD/fmp_cifar10.py","link_basis":"harvester_set","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":"e906d52e487c7b36"}},{"code_sha256_prefix":"33e29a8cc5c2fe60","entry":"lr_decay","repo":"laplacetw/vgg-like-cifar10","repo_kind":"listed","path":"jason_brownlee_cifar10.py","file_url":"https://github.com/laplacetw/vgg-like-cifar10/blob/HEAD/jason_brownlee_cifar10.py","link_basis":"harvester_set","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":"33e29a8cc5c2fe60"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}